Issues in Synchronizing the English Treebank and Propbank

Total Page:16

File Type:pdf, Size:1020Kb

Issues in Synchronizing the English Treebank and Propbank Issues in Synchronizing the English Treebank and PropBank Olga Babko-Malayaa, Ann Biesa, Ann Taylorb, Szuting Yia, Martha Palmerc, Mitch Marcusa, Seth Kulicka and Libin Shena aUniversity of Pennsylvania, bUniversity of York, cUniversity of Colorado {malayao,bies}@ldc.upenn.edu, {szuting,mitch,skulick,libin}@linc.cis.upenn.edu, [email protected], [email protected] semantic role labeling that can be found, and our Abstract plans for reconciling them. We also investigate the sources of the disagreements, which types of The PropBank primarily adds semantic disagreements can be resolved automatically, role labels to the syntactic constituents in which types require manual adjudication, and for the parsed trees of the Treebank. The which types an agreement between syntactic and goal is for automatic semantic role label- semantic representations cannot be reached. ing to be able to use the domain of local- ity of a predicate in order to find its ar- 1.1 Treebank guments. In principle, this is exactly what The Penn Treebank annotates text for syntactic is wanted, but in practice the PropBank structure, including syntactic argument structure annotators often make choices that do not and rough semantic information. Treebank anno- actually conform to the Treebank parses. tation involves two tasks: part-of-speech tagging As a result, the syntactic features ex- and syntactic annotation. tracted by automatic semantic role label- The first task is to provide a part-of-speech tag ing systems are often inconsistent and for every token. Particularly relevant for Prop- contradictory. This paper discusses in de- Bank work, verbs in any form (active, passive, tail the types of mismatches between the gerund, infinitive, etc.) are marked with a verbal syntactic bracketing and the semantic part of speech (VBP, VBN, VBG, VB, etc.). role labeling that can be found, and our (Marcus, et al. 1993; Santorini 1990) plans for reconciling them. The syntactic annotation task consists of marking constituent boundaries, inserting empty 1 Introduction categories (traces of movement, PRO, pro), showing the relationships between constituents The PropBank corpus annotates the entire Penn (argument/adjunct structures), and specifying a Treebank with predicate argument structures by particular subset of adverbial roles. (Marcus, et adding semantic role labels to the syntactic al. 1994; Bies, et al. 1995) constituents of the Penn Treebank. Constituent boundaries are shown through Theoretically, it is straightforward for PropBank syntactic node labels in the trees. In the simplest annotators to locate possible arguments based on case, a node will contain an entire constituent, the syntactic structure given by a parse tree, and complete with any associated arguments or mark the located constituent with its argument modifiers. However, in structures involving syn- label. We would expect a one-to-one mapping tactic movement, sub-constituents may be dis- between syntactic constituents and semantic placed. In these cases, Treebank annotation arguments. However, in practice, PropBank represents the original position with a trace and annotators often make choices that do not shows the relationship as co-indexing. In (1) be- actually conform to the Penn Treebank parses. low, for example, the direct object of entail is The discrepancies between the PropBank and shown with the trace *T*, which is coindexed to the Penn Treebank obstruct the study of the syn- the WHNP node of the question word what. tax and semantics interface and pose an immedi- ate problem to an automatic semantic role label- (SBARQ (WHNP-1 (WP What )) ing system. A semantic role labeling system is (1) (SQ (VBZ does ) trained on many syntactic features extracted from (NP-SBJ (JJ industrial ) the parse trees, and the discrepancies make the (NN emigration )) training data inconsistent and contradictory. In (VP (VB entail) this paper we discuss in detail the types of mis- (NP *T*-1))) matches between the syntactic bracketing and the (. ?)) In (2), the relative clause modifying a journal- tem (temporal –TMP, locative –LOC, manner ist has been separated from that NP by the prepo- –MNR, purpose –PRP, etc.). sitional phrase to al Riyadh, which is an argu- Inside NPs, the argument/adjunct distinction is ment of the verb sent. The position where the shown structurally. Argument constituents (S and relative clause originated or “belongs” is shown SBAR only) are children of NP, sister to the head by the trace *ICH*, which is coindexed to the noun. Adjunct constituents are sister to the NP SBAR node containing the relative clause con- that contains the head noun, child of the NP that stituent. contains both: (2)(S (NP-SBJ You) (NP (NP head) (VP sent (PP adjunct)) (NP (NP a journalist) (SBAR *ICH*-2)) 1.2 PropBank (PP-DIR to (NP al Riyadh)) PropBank is an annotation of predicate-argument (SBAR-2 structures on top of syntactically parsed, or Tree- (WHNP-3 who) banked, structures. (Palmer, et al. 2005; Babko- (S (NP-SBJ *T*-3) Malaya, 2005). More specifically, PropBank (VP served annotation involves three tasks: argument (NP (NP the name) labeling, annotation of modifiers, and creating (PP of co-reference chains for empty categories. (NP Lebanon))) The first goal is to provide consistent argu- (ADVP-MNR magnificently)))))) ment labels across different syntactic realizations of the same verb, as in Empty subjects which are not traces of move- (3) [ARG0 John] broke [ARG1 the window] ment, such as PRO and pro, are shown as * (see [ARG1 The window] broke. the null subject of the infinite clause in (4) be- low). These null subjects are coindexed with a As this example shows, semantic arguments governing NP if the syntax allows. The null sub- are tagged with numbered argument labels, such ject of an infinitive clause complement to a noun as Arg0, Arg1, Arg2, where these labels are de- is, however, not coindexed with another node in fined on a verb by verb basis. the tree in the syntax. This coindexing is shown The second task of the PropBank annotation as a semantic coindexing in the PropBank anno- involves assigning functional tags to all modifi- tation. ers of the verb, such as MNR (manner), LOC The distinction between syntactic arguments (locative), TMP (temporal), DIS (discourse con- and adjuncts of the verb or verb phrase is made nectives), PRP (purpose) or DIR (direction) and through the use of functional dashtags rather than others. with a structural difference. Both arguments and And, finally, PropBank annotation involves adjuncts are children of the VP node. No distinc- finding antecedents for ‘empty’ arguments of the tion is made between VP-level modification and verbs, as in (4). The subject of the verb leave in S-level modification. All constituents that appear this example is represented as an empty category before the verb are children of S and sisters of [*] in Treebank. In PropBank, all empty catego- VP; all constituents that appear after the verb are ries which could be co-referred with a NP within children of VP. the same sentence are linked in ‘co-reference’ Syntactic arguments of the verb are NP-SBJ, chains: NP (no dashtag), SBAR (either –NOM-SBJ or no dashtag), S (either –NOM-SBJ or no dashtag), (4) I made a decision [*] to leave -DTV, -CLR (closely/clearly related), -DIR with directional verbs. Rel: leave, Adjuncts or modifiers of the verb or sentence Arg0: [*] -> I are any constituent with any other adverbial dashtag, PP (no dashtag), ADVP (no dashtag). As the following sections show, all three tasks Adverbial constituents are marked with a more of PropBank annotation result in structures specific functional dashtag if they belong to one which differ in certain respects from the corre- of the more specific types in the annotation sys- sponding Treebank structures. Section 2 presents our approach to reconciling the differences be- (7) What-1 do you like [*T*-1]? tween Treebank and PropBank with respect to the third task, which links empty categories with Revised PropBank annotation: their antecedents. Section 3 introduces mis- Rel: like matches between syntactic constituency in Tree- Arg0: you bank and PropBank. Mismatches between modi- Arg1: [*T*] fier labels are not addressed in this paper and are left for future work. As the second stage, syntactic chains will be re- constructed automatically, based on the 2 Coreference and syntactic chains coindexation provided by Treebank (note that the trace is coindexed with the NP What in (7)). And, PropBank chains include all syntactic chains finally, coreference annotation will be done on (represented in the Treebank) plus other cases of top of the resulting resource, with the goal of nominal semantic coreference, including those finding antecedents for the remaining empty in which the coreferring NP is not a syntactic categories, including empty subjects of infinitival antecedent. For example, according to PropBank verbs and gerunds. guidelines, if a trace is coindexed with a NP in One of the advantages of this approach is that Treebank, then the chain should be reconstructed: it allows us to distinguish different types of chains, such as syntactic chains (i.e., chains (5) What-1 do you like [*T*-1]? which are derived as the result of syntactic movement, or control coreference), direct Original PropBank annotation: coreference chains (as illustrated by the example Rel: like in (6)), and semantic type links for other ‘indi- Arg0: you rect’ types of links between an empty category Arg1: [*T*] -> What and its antecedent. Syntactic chains are annotated in Treebank, Such chains usually include traces of A and A’ and are reconstructed automatically in PropBank. movement and PRO for subject and object con- The annotation of direct coreference chains is trol. On the other hand, not all instances of PROs done manually on top of Treebank, and is re- have syntactic antecedents. As the following ex- stricted to empty categories that are not ample illustrates, subjects of infinitival verbs and coindexed with any NP in Treebank.
Recommended publications
  • How Do BERT Embeddings Organize Linguistic Knowledge?
    How Do BERT Embeddings Organize Linguistic Knowledge? Giovanni Puccettiy , Alessio Miaschi? , Felice Dell’Orletta y Scuola Normale Superiore, Pisa ?Department of Computer Science, University of Pisa Istituto di Linguistica Computazionale “Antonio Zampolli”, Pisa ItaliaNLP Lab – www.italianlp.it [email protected], [email protected], [email protected] Abstract et al., 2019), we proposed an in-depth investigation Several studies investigated the linguistic in- aimed at understanding how the information en- formation implicitly encoded in Neural Lan- coded by BERT is arranged within its internal rep- guage Models. Most of these works focused resentation. In particular, we defined two research on quantifying the amount and type of in- questions, aimed at: (i) investigating the relation- formation available within their internal rep- ship between the sentence-level linguistic knowl- resentations and across their layers. In line edge encoded in a pre-trained version of BERT and with this scenario, we proposed a different the number of individual units involved in the en- study, based on Lasso regression, aimed at understanding how the information encoded coding of such knowledge; (ii) understanding how by BERT sentence-level representations is ar- these sentence-level properties are organized within ranged within its hidden units. Using a suite of the internal representations of BERT, identifying several probing tasks, we showed the existence groups of units more relevant for specific linguistic of a relationship between the implicit knowl- tasks. We defined a suite of probing tasks based on edge learned by the model and the number of a variable selection approach, in order to identify individual units involved in the encodings of which units in the internal representations of BERT this competence.
    [Show full text]
  • Treebanks, Linguistic Theories and Applications Introduction to Treebanks
    Treebanks, Linguistic Theories and Applications Introduction to Treebanks Lecture One Petya Osenova and Kiril Simov Sofia University “St. Kliment Ohridski”, Bulgaria Bulgarian Academy of Sciences, Bulgaria ESSLLI 2018 30th European Summer School in Logic, Language and Information (6 August – 17 August 2018) Plan of the Lecture • Definition of a treebank • The place of the treebank in the language modeling • Related terms: parsebank, dynamic treebank • Prerequisites for the creation of a treebank • Treebank lifecycle • Theory (in)dependency • Language (in)dependency • Tendences in the treebank development 30th European Summer School in Logic, Language and Information (6 August – 17 August 2018) Treebank Definition A corpus annotated with syntactic information • The information in the annotation is added/checked by a trained annotator - manual annotation • The annotation is complete - no unannotated fragments of the text • The annotation is consistent - similar fragments are analysed in the same way • The primary format of annotation - syntactic tree/graph 30th European Summer School in Logic, Language and Information (6 August – 17 August 2018) Example Syntactic Trees from Wikipedia The two main approaches to modeling the syntactic information 30th European Summer School in Logic, Language and Information (6 August – 17 August 2018) Pros vs. Cons (Handbook of NLP, p. 171) Constituency • Easy to read • Correspond to common grammatical knowledge (phrases) • Introduce arbitrary complexity Dependency • Flexible • Also correspond to common grammatical
    [Show full text]
  • Senserelate::Allwords - a Broad Coverage Word Sense Tagger That Maximizes Semantic Relatedness
    WordNet::SenseRelate::AllWords - A Broad Coverage Word Sense Tagger that Maximizes Semantic Relatedness Ted Pedersen and Varada Kolhatkar Department of Computer Science University of Minnesota Duluth, MN 55812 USA tpederse,kolha002 @d.umn.edu http://senserelate.sourceforge.net{ } Abstract Despite these difficulties, word sense disambigua- tion is often a necessary step in NLP and can’t sim- WordNet::SenseRelate::AllWords is a freely ply be ignored. The question arises as to how to de- available open source Perl package that as- velop broad coverage sense disambiguation modules signs a sense to every content word (known that can be deployed in a practical setting without in- to WordNet) in a text. It finds the sense of vesting huge sums in manual annotation efforts. Our each word that is most related to the senses answer is WordNet::SenseRelate::AllWords (SR- of surrounding words, based on measures found in WordNet::Similarity. This method is AW), a method that uses knowledge already avail- shown to be competitive with results from re- able in the lexical database WordNet to assign senses cent evaluations including SENSEVAL-2 and to every content word in text, and as such offers SENSEVAL-3. broad coverage and requires no manual annotation of training data. SR-AW finds the sense of each word that is most 1 Introduction related or most similar to those of its neighbors in the Word sense disambiguation is the task of assigning sentence, according to any of the ten measures avail- a sense to a word based on the context in which it able in WordNet::Similarity (Pedersen et al., 2004).
    [Show full text]
  • Deep Linguistic Analysis for the Accurate Identification of Predicate
    Deep Linguistic Analysis for the Accurate Identification of Predicate-Argument Relations Yusuke Miyao Jun'ichi Tsujii Department of Computer Science Department of Computer Science University of Tokyo University of Tokyo [email protected] CREST, JST [email protected] Abstract obtained by deep linguistic analysis (Gildea and This paper evaluates the accuracy of HPSG Hockenmaier, 2003; Chen and Rambow, 2003). parsing in terms of the identification of They employed a CCG (Steedman, 2000) or LTAG predicate-argument relations. We could directly (Schabes et al., 1988) parser to acquire syntac- compare the output of HPSG parsing with Prop- tic/semantic structures, which would be passed to Bank annotations, by assuming a unique map- statistical classifier as features. That is, they used ping from HPSG semantic representation into deep analysis as a preprocessor to obtain useful fea- PropBank annotation. Even though PropBank tures for training a probabilistic model or statistical was not used for the training of a disambigua- tion model, an HPSG parser achieved the ac- classifier of a semantic argument identifier. These curacy competitive with existing studies on the results imply the superiority of deep linguistic anal- task of identifying PropBank annotations. ysis for this task. Although the statistical approach seems a reason- 1 Introduction able way for developing an accurate identifier of Recently, deep linguistic analysis has successfully PropBank annotations, this study aims at establish- been applied to real-world texts. Several parsers ing a method of directly comparing the outputs of have been implemented in various grammar for- HPSG parsing with the PropBank annotation in or- malisms and empirical evaluation has been re- der to explicitly demonstrate the availability of deep ported: LFG (Riezler et al., 2002; Cahill et al., parsers.
    [Show full text]
  • The Procedure of Lexico-Semantic Annotation of Składnica Treebank
    The Procedure of Lexico-Semantic Annotation of Składnica Treebank Elzbieta˙ Hajnicz Institute of Computer Science, Polish Academy of Sciences ul. Ordona 21, 01-237 Warsaw, Poland [email protected] Abstract In this paper, the procedure of lexico-semantic annotation of Składnica Treebank using Polish WordNet is presented. Other semantically annotated corpora, in particular treebanks, are outlined first. Resources involved in annotation as well as a tool called Semantikon used for it are described. The main part of the paper is the analysis of the applied procedure. It consists of the basic and correction phases. During basic phase all nouns, verbs and adjectives are annotated with wordnet senses. The annotation is performed independently by two linguists. Multi-word units obtain special tags, synonyms and hypernyms are used for senses absent in Polish WordNet. Additionally, each sentence receives its general assessment. During the correction phase, conflicts are resolved by the linguist supervising the process. Finally, some statistics of the results of annotation are given, including inter-annotator agreement. The final resource is represented in XML files preserving the structure of Składnica. Keywords: treebanks, wordnets, semantic annotation, Polish 1. Introduction 2. Semantically Annotated Corpora It is widely acknowledged that linguistically annotated cor- Semantic annotation of text corpora seems to be the last pora play a crucial role in NLP. There is even a tendency phase in the process of corpus annotation, less popular than towards their ever-deeper annotation. In particular, seman- morphosyntactic and (shallow or deep) syntactic annota- tically annotated corpora become more and more popular tion. However, there exist semantically annotated subcor- as they have a number of applications in word sense disam- pora for many languages.
    [Show full text]
  • Unified Language Model Pre-Training for Natural
    Unified Language Model Pre-training for Natural Language Understanding and Generation Li Dong∗ Nan Yang∗ Wenhui Wang∗ Furu Wei∗† Xiaodong Liu Yu Wang Jianfeng Gao Ming Zhou Hsiao-Wuen Hon Microsoft Research {lidong1,nanya,wenwan,fuwei}@microsoft.com {xiaodl,yuwan,jfgao,mingzhou,hon}@microsoft.com Abstract This paper presents a new UNIfied pre-trained Language Model (UNILM) that can be fine-tuned for both natural language understanding and generation tasks. The model is pre-trained using three types of language modeling tasks: unidirec- tional, bidirectional, and sequence-to-sequence prediction. The unified modeling is achieved by employing a shared Transformer network and utilizing specific self-attention masks to control what context the prediction conditions on. UNILM compares favorably with BERT on the GLUE benchmark, and the SQuAD 2.0 and CoQA question answering tasks. Moreover, UNILM achieves new state-of- the-art results on five natural language generation datasets, including improving the CNN/DailyMail abstractive summarization ROUGE-L to 40.51 (2.04 absolute improvement), the Gigaword abstractive summarization ROUGE-L to 35.75 (0.86 absolute improvement), the CoQA generative question answering F1 score to 82.5 (37.1 absolute improvement), the SQuAD question generation BLEU-4 to 22.12 (3.75 absolute improvement), and the DSTC7 document-grounded dialog response generation NIST-4 to 2.67 (human performance is 2.65). The code and pre-trained models are available at https://github.com/microsoft/unilm. 1 Introduction Language model (LM) pre-training has substantially advanced the state of the art across a variety of natural language processing tasks [8, 29, 19, 31, 9, 1].
    [Show full text]
  • Building a Treebank for French
    Building a treebank for French £ £¥ Anne Abeillé£ , Lionel Clément , Alexandra Kinyon ¥ £ TALaNa, Université Paris 7 University of Pennsylvania 75251 Paris cedex 05 Philadelphia FRANCE USA abeille, clement, [email protected] Abstract Very few gold standard annotated corpora are currently available for French. We present an ongoing project to build a reference treebank for French starting with a tagged newspaper corpus of 1 Million words (Abeillé et al., 1998), (Abeillé and Clément, 1999). Similarly to the Penn TreeBank (Marcus et al., 1993), we distinguish an automatic parsing phase followed by a second phase of systematic manual validation and correction. Similarly to the Prague treebank (Hajicova et al., 1998), we rely on several types of morphosyntactic and syntactic annotations for which we define extensive guidelines. Our goal is to provide a theory neutral, surface oriented, error free treebank for French. Similarly to the Negra project (Brants et al., 1999), we annotate both constituents and functional relations. 1. The tagged corpus pronoun (= him ) or a weak clitic pronoun (= to him or to As reported in (Abeillé and Clément, 1999), we present her), plus can either be a negative adverb (= not any more) the general methodology, the automatic tagging phase, the or a simple adverb (= more). Inflectional morphology also human validation phase and the final state of the tagged has to be annotated since morphological endings are impor- corpus. tant for gathering constituants (based on agreement marks) and also because lots of forms in French are ambiguous 1.1. Methodology with respect to mode, person, number or gender. For exam- 1.1.1.
    [Show full text]
  • Converting an HPSG-Based Treebank Into Its Parallel Dependency-Based Treebank
    Converting an HPSG-based Treebank into its Parallel Dependency-based Treebank Masood Ghayoomiy z Jonas Kuhnz yDepartment of Mathematics and Computer Science, Freie Universität Berlin z Institute for Natural Language Processing, University of Stuttgart [email protected] [email protected] Abstract A treebank is an important language resource for supervised statistical parsers. The parser induces the grammatical properties of a language from this language resource and uses the model to parse unseen data automatically. Since developing such a resource is very time-consuming and tedious, one can take advantage of already extant resources by adapting them to a particular application. This reduces the amount of human effort required to develop a new language resource. In this paper, we introduce an algorithm to convert an HPSG-based treebank into its parallel dependency-based treebank. With this converter, we can automatically create a new language resource from an existing treebank developed based on a grammar formalism. Our proposed algorithm is able to create both projective and non-projective dependency trees. Keywords: Treebank Conversion, the Persian Language, HPSG-based Treebank, Dependency-based Treebank 1. Introduction treebank is introduced in Section 4. The properties Supervised statistical parsers require a set of annotated of the available dependency treebanks for Persian and data, called a treebank, to learn the grammar of the their pros and cons are explained in Section 5. The target language and create a wide coverage grammar tools and the experimental setup as well as the ob- model to parse unseen data. Developing such a data tained results are described and discussed in Section source is a tedious and time-consuming task.
    [Show full text]
  • Corpus Based Evaluation of Stemmers
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Repository of the Academy's Library Corpus based evaluation of stemmers Istvan´ Endredy´ Pazm´ any´ Peter´ Catholic University, Faculty of Information Technology and Bionics MTA-PPKE Hungarian Language Technology Research Group 50/a Prater´ Street, 1083 Budapest, Hungary [email protected] Abstract There are many available stemmer modules, and the typical questions are usually: which one should be used, which will help our system and what is its numerical benefit? This article is based on the idea that any corpora with lemmas can be used for stemmer evaluation, not only for lemma quality measurement but for information retrieval quality as well. The sentences of the corpus act as documents (hit items of the retrieval), and its words are the testing queries. This article describes this evaluation method, and gives an up-to-date overview about 10 stemmers for 3 languages with different levels of inflectional morphology. An open source, language independent python tool and a Lucene based stemmer evaluator in java are also provided which can be used for evaluating further languages and stemmers. 1. Introduction 2. Related work Stemmers can be evaluated in several ways. First, their Information retrieval (IR) systems have a common crit- impact on IR performance is measurable by test collec- ical point. It is when the root form of an inflected word tions. The earlier stemmer evaluations (Hull, 1996; Tordai form is determined. This is called stemming. There are and De Rijke, 2006; Halacsy´ and Tron,´ 2007) were based two types of the common base word form: lemma or stem.
    [Show full text]
  • Building a User-Generated Content North-African Arabizi Treebank: Tackling Hell
    Building a User-Generated Content North-African Arabizi Treebank: Tackling Hell Djamé Seddah1 Farah Essaidi1 Amal Fethi1 Matthieu Futeral1 Benjamin Muller1;2 Pedro Javier Ortiz Suárez1;2 Benoît Sagot1 Abhishek Srivastava1 1Inria, Paris, France 2Sorbonne Université, Paris, France [email protected] Abstract sourced, namely Haitian Creole and Algerian di- alectal Arabic in these cases. No readily avail- We introduce the first treebank for a romanized able parsing and machine translations systems are user-generated content variety of Algerian, a available for such languages. Taking as an ex- North-African Arabic dialect known for its fre- quent usage of code-switching. Made of 1500 ample the Arabic dialects spoken in North-Africa, sentences, fully annotated in morpho-syntax mostly from Morocco to Tunisia, sometimes called and Universal Dependency syntax, with full Maghribi, sometimes Darija, these idioms notori- translation at both the word and the sentence ously contain various degrees of code-switching levels, this treebank is made freely available. with languages of former colonial powers such as It is supplemented with 50k unlabeled sen- French, Spanish, and, to a much lesser extent, Ital- tences collected from Common Crawl and web- ian, depending on the area of usage (Habash, 2010; crawled data using intensive data-mining tech- Cotterell et al., 2014; Saadane and Habash, 2015). niques. Preliminary experiments demonstrate its usefulness for POS tagging and dependency They share Modern Standard Arabic (MSA) as parsing. We believe that what we present in their matrix language (Myers-Scotton, 1993), and this paper is useful beyond the low-resource of course present a rich morphology.
    [Show full text]
  • Lecture 5: Part-Of-Speech Tagging
    CS447: Natural Language Processing http://courses.engr.illinois.edu/cs447 Lecture 5: Part-of-Speech Tagging Julia Hockenmaier [email protected] 3324 Siebel Center POS tagging Tagset: NNP: proper noun CD: numeral, POS tagger JJ: adjective, ... Raw text Tagged text Pierre_NNP Vinken_NNP ,_, 61_CD Pierre Vinken , 61 years old years_NNS old_JJ ,_, will_MD join_VB , will join the board as a the_DT board_NN as_IN a_DT nonexecutive director Nov. nonexecutive_JJ director_NN Nov._NNP 29 . 29_CD ._. CS447: Natural Language Processing (J. Hockenmaier) !2 Why POS tagging? POS tagging is a prerequisite for further analysis: –Speech synthesis: How to pronounce “lead”? INsult or inSULT, OBject or obJECT, OVERflow or overFLOW," DIScount or disCOUNT, CONtent or conTENT –Parsing: What words are in the sentence? –Information extraction: Finding names, relations, etc. –Machine Translation: The noun “content” may have a different translation from the adjective. CS447: Natural Language Processing (J. Hockenmaier) !3 POS Tagging Words often have more than one POS: " -The back door (adjective) -On my back (noun) -Win the voters back (particle) -Promised to back the bill (verb)" The POS tagging task is to determine the POS tag " for a particular instance of a word. " Since there is ambiguity, we cannot simply look up the correct POS in a dictionary. These examples from Dekang Lin CS447: Natural Language Processing (J. Hockenmaier) !4 Defining a tagset CS447: Natural Language Processing (J. Hockenmaier) !5 Defining a tag set We have to define an inventory of labels for the word classes (i.e. the tag set)" -Most taggers rely on models that have to be trained on annotated (tagged) corpora.
    [Show full text]
  • Merging Propbank, Nombank, Timebank, Penn Discourse Treebank and Coreference James Pustejovsky, Adam Meyers, Martha Palmer, Massimo Poesio
    Merging PropBank, NomBank, TimeBank, Penn Discourse Treebank and Coreference James Pustejovsky, Adam Meyers, Martha Palmer, Massimo Poesio Abstract annotates the temporal features of propositions Many recent annotation efforts for English and the temporal relations between propositions. have focused on pieces of the larger problem The Penn Discourse Treebank (Miltsakaki et al of semantic annotation, rather than initially 2004a/b) treats discourse connectives as producing a single unified representation. predicates and the sentences being joined as This paper discusses the issues involved in arguments. Researchers at Essex were merging four of these efforts into a unified responsible for the coreference markup scheme linguistic structure: PropBank, NomBank, the developed in MATE (Poesio et al, 1999; Poesio, Discourse Treebank and Coreference 2004a) and have annotated corpora using this Annotation undertaken at the University of scheme including a subset of the Penn Treebank Essex. We discuss resolving overlapping and (Poesio and Vieira, 1998), and the GNOME conflicting annotation as well as how the corpus (Poesio, 2004a). This paper discusses the various annotation schemes can reinforce issues involved in creating a Unified Linguistic each other to produce a representation that is Annotation (ULA) by merging annotation of greater than the sum of its parts. examples using the schemata from these efforts. Crucially, all individual annotations can be kept 1. Introduction separate in order to make it easy to produce alternative annotations of a specific type of The creation of the Penn Treebank (Marcus et al, semantic information without need to modify the 1993) and the word sense-annotated SEMCOR annotation at the other levels. Embarking on (Fellbaum, 1997) have shown how even limited separate annotation efforts has the advantage of amounts of annotated data can result in major allowing researchers to focus on the difficult improvements in complex natural language issues in each area of semantic annotation and understanding systems.
    [Show full text]