Speech and Language Processing. Daniel Jurafsky & James H. Martin. Copyright c© 2015. All
rights reserved. Draft of June 26, 2015.
CHAPTER
22 Semantic Role Labeling
Understanding events and their participants is a key part of understanding natural
language. At a high level, understanding an event means being able to answer the
question “Who did what to whom” (and perhaps also “when and where”). The
answers to this question may be expressed in many different ways in the sentence.
For example, if we want to process sentences to help us answer question about a
purchase of stock by XYZ Corporation, we need to understand this event despite
many different surface forms. The event could be described by a verb (sold, bought)
or a noun (purchase), and XYZ Corp can be the syntactic subject (of bought) the
indirect object (of sold), or in a genitive or noun compound relation (with the noun
purchase), in the following sentences, despite having notationally the same role in
all of them:
• XYZ corporation bought the stock.
• They sold the stock to XYZ corporation.
• The stock was bought by XYZ corporation.
• The purchase of the stock by XYZ corporation…
• The stock purchase by XYZ corporation…
In this chapter we introduce a level of representation that lets us capture the
commonality between these sentences. We will be able to represent the fact that
there was a purchase event, that the participants in this event were XYZ Corp and
some stock, and that XYZ Corp played a specific role, the role of acquiring the stock.
We call this shallow semantic representation level semantic roles. Semantic
roles are representations that express the abstract role that arguments of a predicate
can take in the event; these can be very specific, like the BUYER, abstract like the
AGENT, or super-abstract (the PROTO-AGENT). These roles can both represent gen-
eral semantic properties of the arguments and also express their likely relationship to
the syntactic role of the argument in the sentence. AGENTS tend to be the subject of
an active sentence, THEMES the direct object, and so on. These relations are codified
in databases like PropBank and FrameNet. We’ll introduce semantic role labeling,
the task of assigning roles to the constituents or phrases in sentences. We’ll also
discuss selectional restrictions, the semantic sortal restrictions or preferences that
each individual predicate can express about its potential arguments, such as the fact
that the theme of the verb eat is generally something edible. Along the way, we’ll
describe the various ways these representations can help in language understanding
tasks like question answering and machine translation.
22.1 Semantic Roles
Consider how in Chapter 14 we represented the meaning of arguments for sentences
like these:
2 CHAPTER 22 • SEMANTIC ROLE LABELING
Thematic Role Definition
AGENT The volitional causer of an event
EXPERIENCER The experiencer of an event
FORCE The non-volitional causer of the event
THEME The participant most directly affected by an event
RESULT The end product of an event
CONTENT The proposition or content of a propositional event
INSTRUMENT An instrument used in an event
BENEFICIARY The beneficiary of an event
SOURCE The origin of the object of a transfer event
GOAL The destination of an object of a transfer event
Figure 22.1 Some commonly used thematic roles with their definitions.
(22.1) Sasha broke the window.
(22.2) Pat opened the door.
A neo-Davidsonian event representation of these two sentences would be
∃e,x,y Breaking(e)∧Breaker(e,Sasha)
∧BrokenT hing(e,y)∧Window(y)
∃e,x,y Opening(e)∧Opener(e,Pat)
∧OpenedT hing(e,y)∧Door(y)
In this representation, the roles of the subjects of the verbs break and open are
Breaker and Opener respectively. These deep roles are specific to each event; Break-deep roles
ing events have Breakers, Opening events have Openers, and so on.
If we are going to be able to answer questions, perform inferences, or do any
further kinds of natural language understanding of these events, we’ll need to know
a little more about the semantics of these arguments. Breakers and Openers have
something in common. They are both volitional actors, often animate, and they have
direct causal responsibility for their events.
Thematic roles are a way to capture this semantic commonality between Break-Thematic roles
ers and Eaters.
We say that the subjects of both these verbs are agents. Thus, AGENT is theagents
thematic role that represents an abstract idea such as volitional causation. Similarly,
the direct objects of both these verbs, the BrokenThing and OpenedThing, are both
prototypically inanimate objects that are affected in some way by the action. The
semantic role for these participants is theme.theme
Thematic roles are one of the oldest linguistic models, proposed first by the
Indian grammarian Panini sometime between the 7th and 4th centuries BCE. Their
modern formulation is due to Fillmore (1968) and Gruber (1965). Although there is
no universally agreed-upon set of roles, Figs. 22.1 and 22.2 list some thematic roles
that have been used in various computational papers, together with rough definitions
and examples. Most thematic role sets have about a dozen roles, but we’ll see sets
with smaller numbers of roles with even more abstract meanings, and sets with very
large numbers of roles that are specific to situations. We’ll use the general term
semantic roles for all sets of roles, whether small or large.semantic roles
22.2 • DIATHESIS ALTERNATIONS 3
Thematic Role Example
AGENT The waiter spilled the soup.
EXPERIENCER John has a headache.
FORCE The wind blows debris from the mall into our yards.
THEME Only after Benjamin Franklin broke the ice…
RESULT The city built a regulation-size baseball diamond…
CONTENT Mona asked “You met Mary Ann at a supermarket?”
INSTRUMENT He poached catfish, stunning them with a shocking device…
BENEFICIARY Whenever Ann Callahan makes hotel reservations for her boss…
SOURCE I flew in from Boston.
GOAL I drove to Portland.
Figure 22.2 Some prototypical examples of various thematic roles.
22.2 Diathesis Alternations
The main reason computational systems use semantic roles is to act as a shallow
meaning representation that can let us make simple inferences that aren’t possible
from the pure surface string of words, or even from the parse tree. To extend the
earlier examples, if a document says that Company A acquired Company B, we’d
like to know that this answers the query Was Company B acquired? despite the fact
that the two sentences have very different surface syntax. Similarly, this shallow
semantics might act as a useful intermediate language in machine translation.
Semantic roles thus help generalize over different surface realizations of pred-
icate arguments. For example, while the AGENT is often realized as the subject of
the sentence, in other cases the THEME can be the subject. Consider these possible
realizations of the thematic arguments of the verb break:
(22.3) John
AGENT
broke the window.
THEME
(22.4) John
AGENT
broke the window
THEME
with a rock.
INSTRUMENT
(22.5) The rock
INSTRUMENT
broke the window.
THEME
(22.6) The window
THEME
broke.
(22.7) The window
THEME
was broken by John.
AGENT
These examples suggest that break has (at least) the possible arguments AGENT,
THEME, and INSTRUMENT. The set of thematic role arguments taken by a verb is
often called the thematic grid, θ -grid, or case frame. We can see that there arethematic grid
case frame (among others) the following possibilities for the realization of these arguments of
break:
AGENT/Subject, THEME/Object
AGENT/Subject, THEME/Object, INSTRUMENT/PPwith
INSTRUMENT/Subject, THEME/Object
THEME/Subject
It turns out that many verbs allow their thematic roles to be realized in various
syntactic positions. For example, verbs like give can realize the THEME and GOAL
arguments in two different ways:
4 CHAPTER 22 • SEMANTIC ROLE LABELING
(22.8) a. Doris
AGENT
gave the book
THEME
to Cary.
GOAL
b. Doris
AGENT
gave Cary
GOAL
the book.
THEME
These multiple argument structure realizations (the fact that break can take AGENT,
INSTRUMENT, or THEME as subject, and give can realize its THEME and GOAL in
either order) are called verb alternations or diathesis alternations. The alternationverbalternation
we showed above for give, the dative alternation, seems to occur with particular se-dativealternation
mantic classes of verbs, including “verbs of future having” (advance, allocate, offer,
owe), “send verbs” (forward, hand, mail), “verbs of throwing” (kick, pass, throw),
and so on. Levin (1993) lists for 3100 English verbs the semantic classes to which
they belong (47 high-level classes, divided into 193 more specific classes) and the
various alternations in which they participate. These lists of verb classes have been
incorporated into the online resource VerbNet (Kipper et al., 2000), which links each
verb to both WordNet and FrameNet entries.
22.3 Semantic Roles: Problems with Thematic Roles
Representing meaning at the thematic role level seems like it should be useful in
dealing with complications like diathesis alternations. Yet it has proved quite diffi-
cult to come up with a standard set of roles, and equally difficult to produce a formal
definition of roles like AGENT, THEME, or INSTRUMENT.
For example, researchers attempting to define role sets often find they need to
fragment a role like AGENT or THEME into many specific roles. Levin and Rappa-
port Hovav (2005) summarize a number of such cases, such as the fact there seem
to be at least two kinds of INSTRUMENTS, intermediary instruments that can appear
as subjects and enabling instruments that cannot:
(22.9) a. The cook opened the jar with the new gadget.
b. The new gadget opened the jar.
(22.10) a. Shelly ate the sliced banana with a fork.
b. *The fork ate the sliced banana.
In addition to the fragmentation problem, there are cases in which we’d like to
reason about and generalize across semantic roles, but the finite discrete lists of roles
don’t let us do this.
Finally, it has proved difficult to formally define the thematic roles. Consider the
AGENT role; most cases of AGENTS are animate, volitional, sentient, causal, but any
individual noun phrase might not exhibit all of these properties.
These problems have led to alternative semantic role models that use eithersemantic role
many fewer or many more roles.
The first of these options is to define generalized semantic roles that abstract
over the specific thematic roles. For example, PROTO-AGENT and PROTO-PATIENTproto-agent
proto-patient are generalized roles that express roughly agent-like and roughly patient-like mean-
ings. These roles are defined, not by necessary and sufficient conditions, but rather
by a set of heuristic features that accompany more agent-like or more patient-like
meanings. Thus, the more an argument displays agent-like properties (being voli-
tionally involved in the event, causing an event or a change of state in another par-
ticipant, being sentient or intentionally involved, moving) the greater the likelihood
22.4 • THE PROPOSITION BANK 5
that the argument can be labeled a PROTO-AGENT. The more patient-like the proper-
ties (undergoing change of state, causally affected by another participant, stationary
relative to other participants, etc.), the greater the likelihood that the argument can
be labeled a PROTO-PATIENT.
The second direction is instead to define semantic roles that are specific to a
particular verb or a particular group of semantically related verbs or nouns.
In the next two sections we describe two commonly used lexical resources that
make use of these alternative versions of semantic roles. PropBank uses both proto-
roles and verb-specific semantic roles. FrameNet uses semantic roles that are spe-
cific to a general semantic idea called a frame.
22.4 The Proposition Bank
The Proposition Bank, generally referred to as PropBank, is a resource of sen-PropBank
tences annotated with semantic roles. The English PropBank labels all the sentences
in the Penn TreeBank; the Chinese PropBank labels sentences in the Penn Chinese
TreeBank. Because of the difficulty of defining a universal set of thematic roles,
the semantic roles in PropBank are defined with respect to an individual verb sense.
Each sense of each verb thus has a specific set of roles, which are given only numbers
rather than names: Arg0, Arg1, Arg2, and so on. In general, Arg0 represents the
PROTO-AGENT, and Arg1, the PROTO-PATIENT. The semantics of the other roles
are less consistent, often being defined specifically for each verb. Nonetheless there
are some generalization; the Arg2 is often the benefactive, instrument, attribute, or
end state, the Arg3 the start point, benefactive, instrument, or attribute, and the Arg4
the end point.
Here are some slightly simplified PropBank entries for one sense each of the
verbs agree and fall. Such PropBank entries are called frame files; note that the
definitions in the frame file for each role (“Other entity agreeing”, “Extent, amount
fallen”) are informal glosses intended to be read by humans, rather than being formal
definitions.
(22.11) agree.01
Arg0: Agreer
Arg1: Proposition
Arg2: Other entity agreeing
Ex1: [Arg0 The group] agreed [Arg1 it wouldn’t make an offer].
Ex2: [ArgM-TMP Usually] [Arg0 John] agrees [Arg2 with Mary]
[Arg1 on everything].
(22.12) fall.01
Arg1: Logical subject, patient, thing falling
Arg2: Extent, amount fallen
Arg3: start point
Arg4: end point, end state of arg1
Ex1: [Arg1 Sales] fell [Arg4 to $25 million] [Arg3 from $27 million].
Ex2: [Arg1 The average junk bond] fell [Arg2 by 4.2%].
Note that there is no Arg0 role for fall, because the normal subject of fall is a
PROTO-PATIENT.
6 CHAPTER 22 • SEMANTIC ROLE LABELING
The PropBank semantic roles can be useful in recovering shallow semantic in-
formation about verbal arguments. Consider the verb increase:
(22.13) increase.01 “go up incrementally”
Arg0: causer of increase
Arg1: thing increasing
Arg2: amount increased by, EXT, or MNR
Arg3: start point
Arg4: end point
A PropBank semantic role labeling would allow us to infer the commonality in
the event structures of the following three examples, that is, that in each case Big
Fruit Co. is the AGENT and the price of bananas is the THEME, despite the differing
surface forms.
(22.14) [Arg0 Big Fruit Co. ] increased [Arg1 the price of bananas].
(22.15) [Arg1 The price of bananas] was increased again [Arg0 by Big Fruit Co. ]
(22.16) [Arg1 The price of bananas] increased [Arg2 5%].
PropBank also has a number of non-numbered arguments called ArgMs, (ArgM-
TMP, ArgM-LOC, etc) which represent modification or adjunct meanings. These are
relatively stable across predicates, so aren’t listed with each frame file. Data labeled
with these modifiers can be helpful in training systems to detect temporal, location,
or directional modification across predicates. Some of the ArgM’s include:
TMP when? yesterday evening, now
LOC where? at the museum, in San Francisco
DIR where to/from? down, to Bangkok
MNR how? clearly, with much enthusiasm
PRP/CAU why? because … , in response to the ruling
REC themselves, each other
ADV miscellaneous
PRD secondary predication …ate the meat raw
While PropBank focuses on verbs, a related project, NomBank (Meyers et al.,
2004) adds annotations to noun predicates. For example the noun agreement in
Apple’s agreement with IBM would be labeled with Apple as the Arg0 and IBM as
the Arg2. This allows semantic role labelers to assign labels to arguments of both
verbal and nominal predicates.
22.5 FrameNet
While making inferences about the semantic commonalities across different sen-
tences with increase is useful, it would be even more useful if we could make such
inferences in many more situations, across different verbs, and also between verbs
and nouns. For example, we’d like to extract the similarity among these three sen-
tences:
(22.17) [Arg1 The price of bananas] increased [Arg2 5%].
(22.18) [Arg1 The price of bananas] rose [Arg2 5%].
(22.19) There has been a [Arg2 5%] rise [Arg1 in the price of bananas].
Note that the second example uses the different verb rise, and the third example
uses the noun rather than the verb rise. We’d like a system to recognize that the
22.5 • FRAMENET 7
price of bananas is what went up, and that 5% is the amount it went up, no matter
whether the 5% appears as the object of the verb increased or as a nominal modifier
of the noun rise.
The FrameNet project is another semantic-role-labeling project that attemptsFrameNet
to address just these kinds of problems (Baker et al. 1998, Fillmore et al. 2003,
Fillmore and Baker 2009, Ruppenhofer et al. 2006). Whereas roles in the PropBank
project are specific to an individual verb, roles in the FrameNet project are specific
to a frame.
What is a frame? Consider the following set of words:
reservation, flight, travel, buy, price, cost, fare, rates, meal, plane
There are many individual lexical relations of hyponymy, synonymy, and so on
between many of the words in this list. The resulting set of relations does not,
however, add up to a complete account of how these words are related. They are
clearly all defined with respect to a coherent chunk of common-sense background
information concerning air travel.
We call the holistic background knowledge that unites these words a frame (Fill-frame
more, 1985). The idea that groups of words are defined with respect to some back-
ground information is widespread in artificial intelligence and cognitive science,
where besides frame we see related works like a model (Johnson-Laird, 1983), ormodel
even script (Schank and Abelson, 1977).script
A frame in FrameNet is a background knowledge structure that defines a set of
frame-specific semantic roles, called frame elements, and includes a set of predi-frame elements
cates that use these roles. Each word evokes a frame and profiles some aspect of the
frame and its elements. The FrameNet dataset includes a set of frames and frame
elements, the lexical units associated with each frame, and a set of labeled example
sentences.
For example, the change position on a scale frame is defined as follows:
This frame consists of words that indicate the change of an Item’s posi-
tion on a scale (the Attribute) from a starting point (Initial value) to an
end point (Final value).
Some of the semantic roles (frame elements) in the frame are defined as in
Fig. 22.3. Note that these are separated into core roles, which are frame specific, andCore roles
non-core roles, which are more like the Arg-M arguments in PropBank, expressedNon-core roles
more general properties of time, location, and so on.
Here are some example sentences:
(22.20) [ITEM Oil] rose [ATTRIBUTE in price] [DIFFERENCE by 2%].
(22.21) [ITEM It] has increased [FINAL STATE to having them 1 day a month].
(22.22) [ITEM Microsoft shares] fell [FINAL VALUE to 7 5/8].
(22.23) [ITEM Colon cancer incidence] fell [DIFFERENCE by 50%] [GROUP among
men].
(22.24) a steady increase [INITIAL VALUE from 9.5] [FINAL VALUE to 14.3] [ITEM
in dividends]
(22.25) a [DIFFERENCE 5%] [ITEM dividend] increase…
Note from these example sentences that the frame includes target words like rise,
fall, and increase. In fact, the complete frame consists of the following words:
8 CHAPTER 22 • SEMANTIC ROLE LABELING
Core Roles
ATTRIBUTE The ATTRIBUTE is a scalar property that the ITEM possesses.
DIFFERENCE The distance by which an ITEM changes its position on the scale.
FINAL STATE A description that presents the ITEM’s state after the change in the ATTRIBUTE’s
value as an independent predication.
FINAL VALUE The position on the scale where the ITEM ends up.
INITIAL STATE A description that presents the ITEM’s state before the change in the AT-
TRIBUTE’s value as an independent predication.
INITIAL VALUE The initial position on the scale from which the ITEM moves away.
ITEM The entity that has a position on the scale.
VALUE RANGE A portion of the scale, typically identified by its end points, along which the
values of the ATTRIBUTE fluctuate.
Some Non-Core Roles
DURATION The length of time over which the change takes place.
SPEED The rate of change of the VALUE.
GROUP The GROUP in which an ITEM changes the value of an
ATTRIBUTE in a specified way.
Figure 22.3 The frame elements in the change position on a scale frame from the FrameNet Labelers
Guide (Ruppenhofer et al., 2006).
VERBS: dwindle move soar escalation shift
advance edge mushroom swell explosion tumble
climb explode plummet swing fall
decline fall reach triple fluctuation ADVERBS:
decrease fluctuate rise tumble gain increasingly
diminish gain rocket growth
dip grow shift NOUNS: hike
double increase skyrocket decline increase
drop jump slide decrease rise
FrameNet also codes relationships between frames, allowing frames to inherit
from each other, or representing relations between frames like causation (and gen-
eralizations among frame elements in different frames can be representing by inher-
itance as well). Thus, there is a Cause change of position on a scale frame that is
linked to the Change of position on a scale frame by the cause relation, but that
adds an AGENT role and is used for causative examples such as the following:
(22.26) [AGENT They] raised [ITEM the price of their soda] [DIFFERENCE by 2%].
Together, these two frames would allow an understanding system to extract the
common event semantics of all the verbal and nominal causative and non-causative
usages.
FrameNets have also been developed for many other languages including Span-
ish, German, Japanese, Portuguese, Italian, and Chinese.
22.6 Semantic Role Labeling
Semantic role labeling (sometimes shortened as SRL) is the task of automaticallysemantic rolelabeling
finding the semantic roles of each argument of each predicate in a sentence. Cur-
rent approaches to semantic role labeling are based on supervised machine learning,
often using the FrameNet and PropBank resources to specify what counts as a pred-
icate, define the set of roles used in the task, and provide training and test sets.
22.6 • SEMANTIC ROLE LABELING 9
Recall that the difference between these two models of semantic roles is that
FrameNet (22.27) employs many frame-specific frame elements as roles, while Prop-
Bank (22.28) uses a smaller number of numbered argument labels that can be inter-
preted as verb-specific labels, along with the more general ARGM labels. Some
examples:
(22.27)
[You] can’t [blame] [the program] [for being unable to identify it]
COGNIZER TARGET EVALUEE REASON
(22.28)
[The San Francisco Examiner] issued [a special edition] [yesterday]
ARG0 TARGET ARG1 ARGM-TMP
A simplified semantic role labeling algorithm is sketched in Fig. 22.4. While
there are a large number of algorithms, many of them use some version of the steps
in this algorithm.
Most algorithms, beginning with the very earliest semantic role analyzers (Sim-
mons, 1973), begin by parsing, using broad-coverage parsers to assign a parse to the
input string. Figure 22.5 shows a parse of (22.28) above. The parse is then traversed
to find all words that are predicates.
For each of these predicates, the algorithm examines each node in the parse tree
and decides the semantic role (if any) it plays for this predicate.
This is generally done by supervised classification. Given a labeled training set
such as PropBank or FrameNet, a feature vector is extracted for each node, using
feature templates described in the next subsection.
A 1-of-N classifier is then trained to predict a semantic role for each constituent
given these features, where N is the number of potential semantic roles plus an
extra NONE role for non-role constituents. Most standard classification algorithms
have been used (logistic regression, SVM, etc). Finally, for each test sentence to be
labeled, the classifier is run on each relevant constituent. We give more details of
the algorithm after we discuss features.
function SEMANTICROLELABEL(words) returns labeled tree
parse←PARSE(words)
for each predicate in parse do
for each node in parse do
featurevector←EXTRACTFEATURES(node, predicate, parse)
CLASSIFYNODE(node, featurevector, parse)
Figure 22.4 A generic semantic-role-labeling algorithm. CLASSIFYNODE is a 1-of-N clas-
sifier that assigns a semantic role (or NONE for non-role constituents), trained on labeled data
such as FrameNet or PropBank.
Features for Semantic Role Labeling
A wide variety of features can be used for semantic role labeling. Most systems use
some generalization of the core set of features introduced by Gildea and Jurafsky
(2000). A typical set of basic features are based on the following feature templates
(demonstrated on the NP-SBJ constituent The San Francisco Examiner in Fig. 22.5):
• The governing predicate, in this case the verb issued. The predicate is a cru-
cial feature since labels are defined only with respect to a particular predicate.
• The phrase type of the constituent, in this case, NP (or NP-SBJ). Some se-
mantic roles tend to appear as NPs, others as S or PP, and so on.
10 CHAPTER 22 • SEMANTIC ROLE LABELING
S
NP-SBJ = ARG0 VP
DT NNP NNP NNP
The San Francisco Examiner
VBD = TARGET NP = ARG1 PP-TMP = ARGM-TMP
issued DT JJ NN IN NP
a special edition around NN NP-TMP
noon yesterday
Figure 22.5 Parse tree for a PropBank sentence, showing the PropBank argument labels. The dotted line
shows the path feature NP↑S↓VP↓VBD for ARG0, the NP-SBJ constituent The San Francisco Examiner.
• The headword of the constituent, Examiner. The headword of a constituent
can be computed with standard head rules, such as those given in Chapter 11
in Fig. ??. Certain headwords (e.g., pronouns) place strong constraints on the
possible semantic roles they are likely to fill.
• The headword part of speech of the constituent, NNP.
• The path in the parse tree from the constituent to the predicate. This path is
marked by the dotted line in Fig. 22.5. Following Gildea and Jurafsky (2000),
we can use a simple linear representation of the path, NP↑S↓VP↓VBD. ↑ and
↓ represent upward and downward movement in the tree, respectively. The
path is very useful as a compact representation of many kinds of grammatical
function relationships between the constituent and the predicate.
• The voice of the clause in which the constituent appears, in this case, active
(as contrasted with passive). Passive sentences tend to have strongly different
linkings of semantic roles to surface form than do active ones.
• The binary linear position of the constituent with respect to the predicate,
either before or after.
• The subcategorization of the predicate, the set of expected arguments that
appear in the verb phrase. We can extract this information by using the phrase-
structure rule that expands the immediate parent of the predicate; VP→ VBD
NP PP for the predicate in Fig. 22.5.
• The named entity type of the constituent.
• The first words and the last word of the constituent.
The following feature vector thus represents the first NP in our example (recall
that most observations will have the value NONE rather than, for example, ARG0,
since most constituents in the parse tree will not bear a semantic role):
ARG0: [issued, NP, Examiner, NNP, NP↑S↓VP↓VBD, active, before, VP → NP PP,
ORG, The, Examiner]
Other features are often used in addition, such as sets of n-grams inside the
constituent, or more complex versions of the path features (the upward or downward
halves, or whether particular nodes occur in the path).
It’s also possible to use dependency parses instead of constituency parses as the
basis of features, for example using dependency parse paths instead of constituency
paths.
22.7 • SELECTIONAL RESTRICTIONS 11
Further Issues in Semantic Role Labeling
Instead of training a single-stage classifier, some role-labeling algorithms break
down the classification task for the arguments of a predicate into multiple steps:
1. Pruning: Since only a small number of the constituents in a sentence are
arguments of any given predicate, many systems use simple heuristics to prune
unlikely constituents.
2. Identification: a binary classification of each node as an argument to be la-
beled or a NONE.
3. Classification: a 1-of-N classification of all the constituents that were labeled
as arguments by the previous stage
The separation of identification and classification may lead to better use of fea-
tures (different features may be useful for the two tasks) or to computational effi-
ciency.
The classification algorithm described above classifies each argument separately
(‘locally’), making the simplifying assumption that each argument of a predicate
can be labeled independently. But this is of course not true; there are many kinds
of interactions between arguments that require a more ‘global’ assignment of labels
to constituents. For example, constituents in FrameNet and PropBank are required
to be non-overlapping. Thus a system may incorrectly label two overlapping con-
stituents as arguments. At the very least it needs to decide which of the two is
correct; better would be to use a global criterion to avoid making this mistake. More
significantly, the semantic roles of constituents are not independent; since PropBank
does not allow multiple identical arguments, labeling one constituent as an ARG0
should greatly increase the probability of another constituent being labeled ARG1.
For this reason, many role labeling systems add a fourth step to deal with global
consistency across the labels in a sentence. This fourth step can be implemented
in many ways. The local classifiers can return a list of possible labels associated
with probabilities for each constituent, and a second-pass re-ranking approach can
be used to choose the best consensus label. Integer linear programming (ILP) is
another common way to choose a solution that conforms best to multiple constraints.
The standard evaluation for semantic role labeling is to require that each ar-
gument label must be assigned to the exactly correct word sequence or parse con-
stituent, and then compute precision, recall, and F-measure. Identification and clas-
sification can also be evaluated separately.
Systems for performing automatic semantic role labeling have been applied widely
to improve the state-of-the-art in tasks across NLP like question answering (Shen
and Lapata 2007, Surdeanu et al. 2011) and machine translation (Liu and Gildea 2010,
Lo et al. 2013).
22.7 Selectional Restrictions
We turn in this section to another way to represent facts about the relationship be-
tween predicates and arguments. A selectional restriction is a semantic type con-selectionalrestriction
straint that a verb imposes on the kind of concepts that are allowed to fill its argument
roles. Consider the two meanings associated with the following example:
(22.29) I want to eat someplace nearby.
12 CHAPTER 22 • SEMANTIC ROLE LABELING
There are two possible parses and semantic interpretations for this sentence. In
the sensible interpretation, eat is intransitive and the phrase someplace nearby is
an adjunct that gives the location of the eating event. In the nonsensical speaker-as-
Godzilla interpretation, eat is transitive and the phrase someplace nearby is the direct
object and the THEME of the eating, like the NP Malaysian food in the following
sentences:
(22.30) I want to eat Malaysian food.
How do we know that someplace nearby isn’t the direct object in this sentence?
One useful cue is the semantic fact that the THEME of EATING events tends to be
something that is edible. This restriction placed by the verb eat on the filler of its
THEME argument is a selectional restriction.
Selectional restrictions are associated with senses, not entire lexemes. We can
see this in the following examples of the lexeme serve:
(22.31) The restaurant serves green-lipped mussels.
(22.32) Which airlines serve Denver?
Example (22.31) illustrates the offering-food sense of serve, which ordinarily re-
stricts its THEME to be some kind of food Example (22.32) illustrates the provides a
commercial service to sense of serve, which constrains its THEME to be some type
of appropriate location.
Selectional restrictions vary widely in their specificity. The verb imagine, for
example, imposes strict requirements on its AGENT role (restricting it to humans
and other animate entities) but places very few semantic requirements on its THEME
role. A verb like diagonalize, on the other hand, places a very specific constraint
on the filler of its THEME role: it has to be a matrix, while the arguments of the
adjectives odorless are restricted to concepts that could possess an odor:
(22.33) In rehearsal, I often ask the musicians to imagine a tennis game.
(22.34) Radon is an odorless gas that can’t be detected by human senses.
(22.35) To diagonalize a matrix is to find its eigenvalues.
These examples illustrate that the set of concepts we need to represent selectional
restrictions (being a matrix, being able to possess an odor, etc) is quite open ended.
This distinguishes selectional restrictions from other features for representing lexical
knowledge, like parts-of-speech, which are quite limited in number.
22.7.1 Representing Selectional Restrictions
One way to capture the semantics of selectional restrictions is to use and extend the
event representation of Chapter 14. Recall that the neo-Davidsonian representation
of an event consists of a single variable that stands for the event, a predicate denoting
the kind of event, and variables and relations for the event roles. Ignoring the issue of
the λ -structures and using thematic roles rather than deep event roles, the semantic
contribution of a verb like eat might look like the following:
∃e,x,y Eating(e)∧Agent(e,x)∧T heme(e,y)
With this representation, all we know about y, the filler of the THEME role, is that
it is associated with an Eating event through the Theme relation. To stipulate the
selectional restriction that y must be something edible, we simply add a new term to
that effect:
∃e,x,y Eating(e)∧Agent(e,x)∧T heme(e,y)∧EdibleT hing(y)
22.7 • SELECTIONAL RESTRICTIONS 13
Sense 1
hamburger, beefburger —
(a fried cake of minced beef served on a bun)
=> sandwich
=> snack food
=> dish
=> nutriment, nourishment, nutrition…
=> food, nutrient
=> substance
=> matter
=> physical entity
=> entity
Figure 22.6 Evidence from WordNet that hamburgers are edible.
When a phrase like ate a hamburger is encountered, a semantic analyzer can
form the following kind of representation:
∃e,x,y Eating(e)∧Eater(e,x)∧T heme(e,y)∧EdibleT hing(y)∧Hamburger(y)
This representation is perfectly reasonable since the membership of y in the category
Hamburger is consistent with its membership in the category EdibleThing, assuming
a reasonable set of facts in the knowledge base. Correspondingly, the representation
for a phrase such as ate a takeoff would be ill-formed because membership in an
event-like category such as Takeoff would be inconsistent with membership in the
category EdibleThing.
While this approach adequately captures the semantics of selectional restrictions,
there are two problems with its direct use. First, using FOL to perform the simple
task of enforcing selectional restrictions is overkill. Other, far simpler, formalisms
can do the job with far less computational cost. The second problem is that this
approach presupposes a large, logical knowledge base of facts about the concepts
that make up selectional restrictions. Unfortunately, although such common-sense
knowledge bases are being developed, none currently have the kind of coverage
necessary to the task.
A more practical approach is to state selectional restrictions in terms of WordNet
synsets rather than as logical concepts. Each predicate simply specifies a WordNet
synset as the selectional restriction on each of its arguments. A meaning representa-
tion is well-formed if the role filler word is a hyponym (subordinate) of this synset.
For our ate a hamburger example, for instance, we could set the selectional
restriction on the THEME role of the verb eat to the synset {food, nutrient}, glossed
as any substance that can be metabolized by an animal to give energy and build
tissue. Luckily, the chain of hypernyms for hamburger shown in Fig. 22.6 reveals
that hamburgers are indeed food. Again, the filler of a role need not match the
restriction synset exactly; it just needs to have the synset as one of its superordinates.
We can apply this approach to the THEME roles of the verbs imagine, lift, and di-
agonalize, discussed earlier. Let us restrict imagine’s THEME to the synset {entity},
lift’s THEME to {physical entity}, and diagonalize to {matrix}. This arrangement
correctly permits imagine a hamburger and lift a hamburger, while also correctly
ruling out diagonalize a hamburger.
14 CHAPTER 22 • SEMANTIC ROLE LABELING
22.7.2 Selectional Preferences
In the earliest implementations, selectional restrictions were considered strict con-
straints on the kind of arguments a predicate could take (Katz and Fodor 1963,
Hirst 1987). For example, the verb eat might require that its THEME argument
be [+FOOD]. Early word sense disambiguation systems used this idea to rule out
senses that violated the selectional restrictions of their governing predicates.
Very quickly, however, it became clear that these selectional restrictions were
better represented as preferences rather than strict constraints (Wilks 1975b, Wilks 1975a).
For example, selectional restriction violations (like inedible arguments of eat) often
occur in well-formed sentences, for example because they are negated (22.36), or
because selectional restrictions are overstated (22.37):
(22.36) But it fell apart in 1931, perhaps because people realized you can’t eat
gold for lunch if you’re hungry.
(22.37) In his two championship trials, Mr. Kulkarni ate glass on an empty
stomach, accompanied only by water and tea.
Modern systems for selectional preferences therefore specify the relation be-
tween a predicate and its possible arguments with soft constraints of some kind.
Selectional Association
One of the most influential has been the selectional association model of Resnik
(1993). Resnik defines the idea of selectional preference strength as the general
selectional
preference
strength
amount of information that a predicate tells us about the semantic class of its argu-
ments. For example, the verb eat tells us a lot about the semantic class of its direct
objects, since they tend to be edible. The verb be, by contrast, tells us less about
its direct objects. The selectional preference strength can be defined by the differ-
ence in information between two distributions: the distribution of expected semantic
classes P(c) (how likely is it that a direct object will fall into class c) and the dis-
tribution of expected semantic classes for the particular verb P(c|v) (how likely is
it that the direct object of the specific verb v will fall into semantic class c). The
greater the difference between these distributions, the more information the verb is
giving us about possible objects. The difference between these two distributions can
be quantified by relative entropy, or the Kullback-Leibler divergence (Kullback andrelative entropy
Leibler, 1951). The Kullback-Leibler or KL divergence D(P||Q) expresses the dif-KL divergence
ference between two probability distributions P and Q (we’ll return to this when we
discuss distributional models of meaning in Chapter 17).
D(P||Q) =
∑
x
P(x) log
P(x)
Q(x)
(22.38)
The selectional preference SR(v) uses the KL divergence to express how much
information, in bits, the verb v expresses about the possible semantic class of its
argument.
SR(v) = D(P(c|v)||P(c))
=
∑
c
P(c|v) log
P(c|v)
P(c)
(22.39)
Resnik then defines the selectional association of a particular class and verbselectionalassociation
22.7 • SELECTIONAL RESTRICTIONS 15
as the relative contribution of that class to the general selectional preference of the
verb:
AR(v,c) =
1
SR(v)
P(c|v) log
P(c|v)
P(c)
(22.40)
The selectional association is thus a probabilistic measure of the strength of as-
sociation between a predicate and a class dominating the argument to the predicate.
Resnik estimates the probabilities for these associations by parsing a corpus, count-
ing all the times each predicate occurs with each argument word, and assuming
that each word is a partial observation of all the WordNet concepts containing the
word. The following table from Resnik (1996) shows some sample high and low
selectional associations for verbs and some WordNet semantic classes of their direct
objects.
Direct Object Direct Object
Verb Semantic Class Assoc Semantic Class Assoc
read WRITING 6.80 ACTIVITY -.20
write WRITING 7.26 COMMERCE 0
see ENTITY 5.79 METHOD -0.01
Selectional Preference via Conditional Probability
An alternative to using selectional association between a verb and the WordNet class
of its arguments, is to simply use the conditional probability of an argument word
given a predicate verb. This simple model of selectional preferences can be used to
directly modeling the strength of association of one verb (predicate) with one noun
(argument).
The conditional probability model can be computed by parsing a very large cor-
pus (billions of words), and computing co-occurrence counts: how often a given
verb occurs with a given noun in a given relation. The conditional probability of an
argument noun given a verb for a particular relation P(n|v,r) can then be used as a
selectional preference metric for that pair of words (Brockmann and Lapata, 2003):
P(n|v,r) =
{
C(n,v,r)
C(v,r) if C(n,v,r)> 0
0 otherwise
The inverse probability P(v|n,r) was found to have better performance in some
cases (Brockmann and Lapata, 2003):
P(v|n,r) =
{
C(n,v,r)
C(n,r) if C(n,v,r)> 0
0 otherwise
In cases where it’s not possible to get large amounts of parsed data, another
option, at least for direct objects, is to get the counts from simple part-of-speech
based approximations. For example pairs can be extracted using the pattern ”V Det
N”, where V is any form of the verb, Det is the—a—ε and N is the singular or plural
form of the noun (Keller and Lapata, 2003).
An even simpler approach is to use the simple log co-occurrence frequency of
the predicate with the argument logcount(v,n,r) instead of conditional probability;
this seems to do better for extracting preferences for syntactic subjects rather than
objects (Brockmann and Lapata, 2003).
16 CHAPTER 22 • SEMANTIC ROLE LABELING
Evaluating Selectional Preferences
One way to evaluate models of selectional preferences is to use pseudowords (Galepseudowords
et al. 1992, Schütze 1992). A pseudoword is an artificial word created by concate-
nating a test word in some context (say banana) with a confounder word (say door)
to create banana-door). The task of the system is to identify which of the two words
is the original word. To evaluate a selectional preference model (for example on the
relationship between a verb and a direct object) we take a test corpus and select all
verb tokens. For each verb token (say drive) we select the direct object (e.g., car),
concatenated with a confounder word that is its nearest neighbor, the noun with the
frequency closest to the original (say house), to make car/house). We then use the
selectional preference model to choose which of car and house are more preferred
objects of drive, and compute how often the model chooses the correct original ob-
ject (e.g., (car) (Chambers and Jurafsky, 2010).
Another evaluation metric is to get human preferences for a test set of verb-
argument pairs, and have them rate their degree of plausibility. This is usually done
by using magnitude estimation, a technique from psychophysics, in which subjects
rate the plausibility of an argument proportional to a modulus item. A selectional
preference model can then be evaluated by its correlation with the human prefer-
ences (Keller and Lapata, 2003).
22.8 Primitive Decomposition of Predicates
One way of thinking about the semantic roles we have discussed through the chapter
is that they help us define the roles that arguments play in a decompositional way,
based on finite lists of thematic roles (agent, patient, instrument, proto-agent, proto-
patient, etc.) This idea of decomposing meaning into sets of primitive semantics
elements or features, called primitive decomposition or componential analysis,componentialanalysis
has been taken even further, and focused particularly on predicates.
Consider these examples of the verb kill:
(22.41) Jim killed his philodendron.
(22.42) Jim did something to cause his philodendron to become not alive.
There is a truth-conditional (‘propositional semantics’) perspective from which these
two sentences have the same meaning. Assuming this equivalence, we could repre-
sent the meaning of kill as:
(22.43) KILL(x,y)⇔ CAUSE(x, BECOME(NOT(ALIVE(y))))
thus using semantic primitives like do, cause, become not, and alive.
Indeed, one such set of potential semantic primitives has been used to account
for some of the verbal alternations discussed in Section 22.2 (Lakoff, 1965; Dowty,
1979). Consider the following examples.
(22.44) John opened the door. ⇒ CAUSE(John(BECOME(OPEN(door))))
(22.45) The door opened. ⇒ BECOME(OPEN(door))
(22.46) The door is open. ⇒ OPEN(door)
The decompositional approach asserts that a single state-like predicate associ-
ated with open underlies all of these examples. The differences among the meanings
of these examples arises from the combination of this single predicate with the prim-
itives CAUSE and BECOME.
22.9 • AMR 17
While this approach to primitive decomposition can explain the similarity be-
tween states and actions or causative and non-causative predicates, it still relies on
having a large number of predicates like open. More radical approaches choose to
break down these predicates as well. One such approach to verbal predicate de-
composition that played a role in early natural language understanding systems is
conceptual dependency (CD), a set of ten primitive predicates, shown in Fig. 22.7.conceptualdependency
Primitive Definition
ATRANS The abstract transfer of possession or control from one entity to
another
PTRANS The physical transfer of an object from one location to another
MTRANS The transfer of mental concepts between entities or within an
entity
MBUILD The creation of new information within an entity
PROPEL The application of physical force to move an object
MOVE The integral movement of a body part by an animal
INGEST The taking in of a substance by an animal
EXPEL The expulsion of something from an animal
SPEAK The action of producing a sound
ATTEND The action of focusing a sense organ
Figure 22.7 A set of conceptual dependency primitives.
Below is an example sentence along with its CD representation. The verb brought
is translated into the two primitives ATRANS and PTRANS to indicate that the waiter
both physically conveyed the check to Mary and passed control of it to her. Note
that CD also associates a fixed set of thematic roles with each primitive to represent
the various participants in the action.
(22.47) The waiter brought Mary the check.
∃x,y Atrans(x)∧Actor(x,Waiter)∧Ob ject(x,Check)∧To(x,Mary)
∧Ptrans(y)∧Actor(y,Waiter)∧Ob ject(y,Check)∧To(y,Mary)
22.9 AMR
To be written
22.10 Summary
• Semantic roles are abstract models of the role an argument plays in the event
described by the predicate.
• Thematic roles are a model of semantic roles based on a single finite list of
roles. Other semantic role models include per-verb semantic role lists and
proto-agent/proto-patient, both of which are implemented in PropBank,
and per-frame role lists, implemented in FrameNet.
18 CHAPTER 22 • SEMANTIC ROLE LABELING
• Semantic role labeling is the task of assigning semantic role labels to the con-
stituents of a sentence. The task is generally treated as a supervised machine
learning task, with models trained on PropBank or FrameNet. Algorithms
generally start by parsing a sentence and then automatically tag each parse
tree node with a semantic role.
• Semantic selectional restrictions allow words (particularly predicates) to post
constraints on the semantic properties of their argument words. Selectional
preference models (like selectional association or simple conditional proba-
bility) allow a weight or probability to be assigned to the association between
a predicate and an argument word or class.
Bibliographical and Historical Notes
Although the idea of semantic roles dates back to Panini, they were re-introduced
into modern linguistics by (Gruber, 1965) and (Fillmore, 1966) and (Fillmore, 1968).
Fillmore, interestingly, had become interested in argument structure by studying
Lucien Tesnière’s groundbreaking Éléments de Syntaxe Structurale (Tesnière, 1959)
in which the term ‘dependency’ was introduced and the foundations were laid for
dependency grammar. Following Tesnière’s terminology, Fillmore first referred to
argument roles as actants (Fillmore, 1966) but quickly switched to the term case,
(see Fillmore (2003)) and proposed a universal list of semantic roles or cases (Agent,
Patient, Instrument, etc.), that could be taken on by the arguments of predicates.
Verbs would be listed in the lexicon with their ‘case frame’, the list of obligatory (or
optional) case arguments.
The idea that semantic roles could provide an intermediate level of semantic
representation that could help map from syntactic parse structures to deeper, more
fully-specified representations of meaning was quickly adopted in natural language
processing, and systems for extracting case frames were created for machine trans-
lation (Wilks, 1973), question-answering (Hendrix et al., 1973), spoken-language
understanding (Nash-Webber, 1975), and dialogue systems (Bobrow et al., 1977).
General-purpose semantic role labelers were developed. The earliest ones (Sim-
mons, 1973) first parsed a sentence by means of an ATN parser. Each verb then had
a set of rules specifying how the parse should be mapped to semantic roles. These
rules mainly made reference to grammatical functions (subject, object, complement
of specific prepositions) but also checked constituent internal features such as the an-
imacy of head nouns. Later systems assigned roles from pre-built parse trees, again
by using dictionaries with verb-specific case frames (Levin 1977, Marcus 1980).
By 1977 case representation was widely used and taught in natural language
processing and artificial intelligence, and was described as a standard component
of natural language understanding in the first edition of Winston’s (1977) textbook
Artificial Intelligence.
In the 1980s Fillmore proposed his model of frame semantics, later describing
the intuition as follows:
“The idea behind frame semantics is that speakers are aware of possi-
bly quite complex situation types, packages of connected expectations,
that go by various names—frames, schemas, scenarios, scripts, cultural
narratives, memes—and the words in our language are understood with
such frames as their presupposed background.” (Fillmore, 2012, p. 712)
EXERCISES 19
The word frame seemed to be in the air for a suite of related notions proposed at
about the same time by Minsky (1974), Hymes (1974), and Goffman (1974), as
well as related notions with other names like scripts (Schank and Abelson, 1975)
and schemata (Bobrow and Norman, 1975) (see Tannen (1979) for a comparison).
Fillmore was also influenced by the semantic field theorists and by a visit to the Yale
AI lab where he took notice of the lists of slots and fillers used by early information
extraction systems like DeJong (1982) and Schank and Abelson (1977). In the 1990s
Fillmore drew on these insights to begin the FrameNet corpus annotation project.
At the same time, Beth Levin drew on her early case frame dictionaries (Levin,
1977) to develop her book which summarized sets of verb classes defined by shared
argument realizations (Levin, 1993). The VerbNet project built on this work (Kipper
et al., 2000), leading soon afterwards to the PropBank semantic-role-labeled corpus
created by Martha Palmer and colleagues (Palmer et al., 2005). The combination of
rich linguistic annotation and corpus-based approach instantiated in FrameNet and
PropBank led to a revival of automatic approaches to semantic role labeling, first
on FrameNet (Gildea and Jurafsky, 2000) and then on PropBank data (Gildea and
Palmer, 2002, inter alia). The problem first addressed in the 1970s by hand-written
rules was thus now generally recast as one of supervised machine learning enabled
by large and consistent databases. Many popular features used for role labeling
are defined in Gildea and Jurafsky (2002), Surdeanu et al. (2003), Xue and Palmer
(2004), Pradhan et al. (2005), Che et al. (2009), and Zhao et al. (2009).
The use of dependency rather than constituency parses was introduced in the
CoNLL-2008 shared task (Surdeanu et al., 2008). For surveys see Palmer et al.
(2010) and Màrquez et al. (2008).
To avoid the need for huge labeled training sets, unsupervised approaches for se-
mantic role labeling attempt to induce the set of semantic roles by generalizing over
syntactic features of arguments (Swier and Stevenson 2004, Grenager and Man-
ning 2006, Titov and Klementiev 2012, Lang and Lapata 2014).
The most recent work in semantic role labeling focuses on the use of deep neural
networks (Collobert et al. 2011, Foland Jr and Martin 2015).
Selectional preference has been widely studied beyond the selectional associa-
tion models of Resnik (1993) and Resnik (1996). Methods have included cluster-
ing (Rooth et al., 1999), discriminative learning (Bergsma et al., 2008), and topic
models (Séaghdha 2010, Ritter et al. 2010), and constraints can be expressed at the
level of words or classes (Agirre and Martinez, 2001). Selectional preferences have
also been successfully integrated into semantic role labeling (Erk 2007, Zapirain
et al. 2013).
Exercises
20 Chapter 22 • Semantic Role Labeling
Agirre, E. and Martinez, D. (2001). Learning class-to-class
selectional preferences. In CoNLL-01.
Baker, C. F., Fillmore, C. J., and Lowe, J. B. (1998). The
Berkeley FrameNet project. In COLING/ACL-98, Mon-
treal, Canada, pp. 86–90.
Bergsma, S., Lin, D., and Goebel, R. (2008). Discriminative
learning of selectional preference from unlabeled text. In
EMNLP-08, pp. 59–68.
Bobrow, D. G., Kaplan, R. M., Kay, M., Norman, D. A.,
Thompson, H., and Winograd, T. (1977). GUS, A frame
driven dialog system. Artificial Intelligence, 8, 155–173.
Bobrow, D. G. and Norman, D. A. (1975). Some princi-
ples of memory schemata. In Bobrow, D. G. and Collins,
A. (Eds.), Representation and Understanding. Academic
Press.
Brockmann, C. and Lapata, M. (2003). Evaluating and com-
bining approaches to selectional preference acquisition. In
EACL-03, pp. 27–34.
Chambers, N. and Jurafsky, D. (2010). Improving the use
of pseudo-words for evaluating selectional preferences. In
ACL 2010, pp. 445–453.
Che, W., Li, Z., Li, Y., Guo, Y., Qin, B., and Liu, T.
(2009). Multilingual dependency-based syntactic and se-
mantic parsing. In CoNLL-09, pp. 49–54.
Collobert, R., Weston, J., Bottou, L., Karlen, M.,
Kavukcuoglu, K., and Kuksa, P. (2011). Natural language
processing (almost) from scratch. The Journal of Machine
Learning Research, 12, 2493–2537.
DeJong, G. F. (1982). An overview of the FRUMP system.
In Lehnert, W. G. and Ringle, M. H. (Eds.), Strategies for
Natural Language Processing, pp. 149–176. Lawrence Erl-
baum.
Dowty, D. R. (1979). Word Meaning and Montague Gram-
mar. D. Reidel.
Erk, K. (2007). A simple, similarity-based model for selec-
tional preferences. In ACL-07, pp. 216–223.
Fillmore, C. J. (1966). A proposal concerning english prepo-
sitions. In Dinneen, F. P. (Ed.), 17th annual Round Table.,
Vol. 17 of Monograph Series on Language and Linguistics,
pp. 19–34. Georgetown University Press, Washington D.C.
Fillmore, C. J. (1968). The case for case. In Bach, E. W. and
Harms, R. T. (Eds.), Universals in Linguistic Theory, pp.
1–88. Holt, Rinehart & Winston.
Fillmore, C. J. (1985). Frames and the semantics of under-
standing. Quaderni di Semantica, VI(2), 222–254.
Fillmore, C. J. (2003). Valency and semantic roles: the con-
cept of deep structure case. In Ágel, V., Eichinger, L. M.,
Eroms, H. W., Hellwig, P., Heringer, H. J., and Lobin, H.
(Eds.), Dependenz und Valenz: Ein internationales Hand-
buch der zeitgenössischen Forschung, chap. 36, pp. 457–
475. Walter de Gruyter.
Fillmore, C. J. (2012). Encounters with language. Computa-
tional Linguistics, 38(4), 701–718.
Fillmore, C. J. and Baker, C. F. (2009). A frames approach
to semantic analysis. In Heine, B. and Narrog, H. (Eds.),
The Oxford Handbook of Linguistic Analysis, pp. 313–340.
Oxford University Press.
Fillmore, C. J., Johnson, C. R., and Petruck, M. R. L. (2003).
Background to FrameNet. International journal of lexicog-
raphy, 16(3), 235–250.
Foland Jr, W. R. and Martin, J. H. (2015). Dependency-
based semantic role labeling using convolutional neural
networks. In *SEM 2015), pp. 279–289.
Gale, W. A., Church, K. W., and Yarowsky, D. (1992). Work
on statistical methods for word sense disambiguation. In
Goldman, R. (Ed.), Proceedings of the 1992 AAAI Fall
Symposium on Probabilistic Approaches to Natural Lan-
guage.
Gildea, D. and Jurafsky, D. (2000). Automatic labeling of
semantic roles. In ACL-00, Hong Kong, pp. 512–520.
Gildea, D. and Jurafsky, D. (2002). Automatic labeling of se-
mantic roles. Computational Linguistics, 28(3), 245–288.
Gildea, D. and Palmer, M. (2002). The necessity of syntac-
tic parsing for predicate argument recognition. In ACL-02,
Philadelphia, PA.
Goffman, E. (1974). Frame analysis: An essay on the orga-
nization of experience. Harvard University Press.
Grenager, T. and Manning, C. D. (2006). Unsupervised Dis-
covery of a Statistical Verb Lexicon. In EMNLP 2006.
Gruber, J. S. (1965). Studies in Lexical Relations. Ph.D.
thesis, MIT.
Hendrix, G. G., Thompson, C. W., and Slocum, J. (1973).
Language processing via canonical verbs and semantic
models. In Proceedings of IJCAI-73.
Hirst, G. (1987). Semantic Interpretation and the Resolution
of Ambiguity. Cambridge University Press.
Hymes, D. (1974). Ways of speaking. In Bauman, R.
and Sherzer, J. (Eds.), Explorations in the ethnography of
speaking, pp. 433–451. Cambridge University Press.
Johnson-Laird, P. N. (1983). Mental Models. Harvard Uni-
versity Press, Cambridge, MA.
Katz, J. J. and Fodor, J. A. (1963). The structure of a seman-
tic theory. Language, 39, 170–210.
Keller, F. and Lapata, M. (2003). Using the web to obtain fre-
quencies for unseen bigrams. Computational Linguistics,
29, 459–484.
Kipper, K., Dang, H. T., and Palmer, M. (2000). Class-based
construction of a verb lexicon. In AAAI-00, Austin, TX, pp.
691–696.
Kullback, S. and Leibler, R. A. (1951). On information and
sufficiency. Annals of Mathematical Statistics, 22, 79–86.
Lakoff, G. (1965). On the Nature of Syntactic Irregularity.
Ph.D. thesis, Indiana University. Published as Irregularity
in Syntax. Holt, Rinehart, and Winston, New York, 1970.
Lang, J. and Lapata, M. (2014). Similarity-driven semantic
role induction via graph partitioning. Computational Lin-
guistics, 40(3), 633–669.
Levin, B. (1977). Mapping sentences to case frames. Tech.
rep. 167, MIT AI Laboratory. AI Working Paper 143.
Levin, B. (1993). English Verb Classes and Alternations: A
Preliminary Investigation. University of Chicago Press.
Levin, B. and Rappaport Hovav, M. (2005). Argument Real-
ization. Cambridge University Press.
Liu, D. and Gildea, D. (2010). Semantic role features for
machine translation. In Proceedings of COLING 2010, pp.
716–724.
Exercises 21
Lo, C.-k., Addanki, K., Saers, M., and Wu, D. (2013). Im-
proving machine translation by training against an auto-
matic semantic frame based evaluation metric. In Proceed-
ings of ACL 2013.
Marcus, M. P. (1980). A Theory of Syntactic Recognition for
Natural Language. MIT Press.
Màrquez, L., Carreras, X., Litkowski, K. C., and Stevenson,
S. (2008). Semantic role labeling: An introduction to the
special issue. Computational linguistics, 34(2), 145–159.
Meyers, A., Reeves, R., Macleod, C., Szekely, R., Zielin-
ska, V., Young, B., and Grishman, R. (2004). The nom-
bank project: An interim report. In Proceedings of the
NAACL/HLT Workshop: Frontiers in Corpus Annotation.
Minsky, M. (1974). A framework for representing knowl-
edge. Tech. rep. 306, MIT AI Laboratory. Memo 306.
Nash-Webber, B. L. (1975). The role of semantics in auto-
matic speech understanding. In Bobrow, D. G. and Collins,
A. (Eds.), Representation and Understanding, pp. 351–
382. Academic Press.
Palmer, M., Gildea, D., and Xue, N. (2010). Semantic role
labeling. Synthesis Lectures on Human Language Tech-
nologies, 3(1), 1–103.
Palmer, M., Kingsbury, P., and Gildea, D. (2005). The propo-
sition bank: An annotated corpus of semantic roles. Com-
putational Linguistics, 31(1), 71–106.
Pradhan, S., Ward, W., Hacioglu, K., Martin, J. H., and Ju-
rafsky, D. (2005). Semantic role labeling using different
syntactic views. In ACL-05, Ann Arbor, MI.
Resnik, P. (1993). Semantic classes and syntactic ambigu-
ity. In Proceedings of the workshop on Human Language
Technology, pp. 278–283.
Resnik, P. (1996). Selectional constraints: An information-
theoretic model and its computational realization. Cogni-
tion, 61, 127–159.
Ritter, A., Etzioni, O., and Mausam (2010). A latent dirich-
let allocation method for selectional preferences. In ACL
2010, pp. 424–434.
Rooth, M., Riezler, S., Prescher, D., Carroll, G., and Beil,
F. (1999). Inducing a semantically annotated lexicon via
EM-based clustering. In ACL-99, College Park, MA, pp.
104–111.
Ruppenhofer, J., Ellsworth, M., Petruck, M. R. L., John-
son, C. R., and Scheffczyk, J. (2006). FrameNet II: Ex-
tended theory and practice. Version 1.3, http://www.
icsi.berkeley.edu/framenet/.
Schank, R. C. and Abelson, R. P. (1975). Scripts, plans, and
knowledge. In Proceedings of IJCAI-75, pp. 151–157.
Schank, R. C. and Abelson, R. P. (1977). Scripts, Plans,
Goals and Understanding. Lawrence Erlbaum.
Schütze, H. (1992). Context space. In Goldman, R. (Ed.),
Proceedings of the 1992 AAAI Fall Symposium on Proba-
bilistic Approaches to Natural Language.
Séaghdha, D. O. (2010). Latent variable models of selec-
tional preference. In ACL 2010, pp. 435–444.
Shen, D. and Lapata, M. (2007). Using semantic roles to im-
prove question answering.. In EMNLP-CoNLL, pp. 12–21.
Simmons, R. F. (1973). Semantic networks: Their com-
putation and use for understanding English sentences. In
Schank, R. C. and Colby, K. M. (Eds.), Computer Models
of Thought and Language, pp. 61–113. W.H. Freeman and
Co.
Surdeanu, M., Ciaramita, M., and Zaragoza, H. (2011).
Learning to rank answers to non-factoid questions from
web collections. Computational Linguistics, 37(2), 351–
383.
Surdeanu, M., Harabagiu, S., Williams, J., and Aarseth, P.
(2003). Using predicate-argument structures for informa-
tion extraction. In ACL-03, pp. 8–15.
Surdeanu, M., Johansson, R., Meyers, A., Màrquez, L., and
Nivre, J. (2008). The conll-2008 shared task on joint pars-
ing of syntactic and semantic dependencies. In CoNLL-08,
pp. 159–177.
Swier, R. and Stevenson, S. (2004). Unsupervised semantic
role labelling. In EMNLP 2004, pp. 95–102.
Tannen, D. (1979). What’s in a frame? Surface evidence for
underlying expectations. In Freedle, R. (Ed.), New Direc-
tions in Discourse Processing, pp. 137–181. Ablex.
Tesnière, L. (1959). Éléments de Syntaxe Structurale. Li-
brairie C. Klincksieck, Paris.
Titov, I. and Klementiev, A. (2012). A Bayesian approach
to unsupervised semantic role induction. In EACL-12, pp.
12–22.
Wilks, Y. (1973). An artificial intelligence approach to ma-
chine translation. In Schank, R. C. and Colby, K. M. (Eds.),
Computer Models of Thought and Language, pp. 114–151.
W.H. Freeman.
Wilks, Y. (1975a). Preference semantics. In Keenan, E. L.
(Ed.), The Formal Semantics of Natural Language, pp.
329–350. Cambridge Univ. Press.
Wilks, Y. (1975b). A preferential, pattern-seeking, semantics
for natural language inference. Artificial Intelligence, 6(1),
53–74.
Winston, P. H. (1977). Artificial Intelligence. Addison Wes-
ley.
Xue, N. and Palmer, M. (2004). Calibrating features for se-
mantic role labeling. In EMNLP 2004.
Zapirain, B., Agirre, E., Màrquez, L., and Surdeanu, M.
(2013). Selectional preferences for semantic role classi-
fication. Computational Linguistics, 39(3), 631–663.
Zhao, H., Chen, W., Kit, C., and Zhou, G. (2009). Mul-
tilingual dependency learning: A huge feature engineering
method to semantic dependency parsing. In CoNLL-09, pp.
55–60.
http://www.icsi.berkeley.edu/framenet/
http://www.icsi.berkeley.edu/framenet/
Semantic Role Labeling
Semantic Roles
Diathesis Alternations
Semantic Roles: Problems with Thematic Roles
The Proposition Bank
FrameNet
Semantic Role Labeling
Selectional Restrictions
Representing Selectional Restrictions
Selectional Preferences
Primitive Decomposition of Predicates
AMR
Summary
Bibliographical and Historical Notes
Exercises