| { | |
| "paper_id": "W14-0139", | |
| "header": { | |
| "generated_with": "S2ORC 1.0.0", | |
| "date_generated": "2023-01-19T05:46:59.706443Z" | |
| }, | |
| "title": "Leveraging Morpho-semantics for the Discovery of Relations in Chinese Wordnet", | |
| "authors": [ | |
| { | |
| "first": "Shu-Kai", | |
| "middle": [], | |
| "last": "Hsieh", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "Graduate Institute of Linguistics National Taiwan University Taipei", | |
| "location": { | |
| "country": "Taiwan" | |
| } | |
| }, | |
| "email": "[email protected]" | |
| }, | |
| { | |
| "first": "Yu-Yun", | |
| "middle": [], | |
| "last": "Chang", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "National Taiwan University Taipei", | |
| "location": { | |
| "country": "Taiwan" | |
| } | |
| }, | |
| "email": "" | |
| } | |
| ], | |
| "year": "", | |
| "venue": null, | |
| "identifiers": {}, | |
| "abstract": "Semantic relations of different types have played an important role in wordnet, and have been widely recognized in various fields. In recent years, with the growing interests of constructing semantic network in support of intelligent systems, automatic semantic relation discovery has become an urgent task. This paper aims to extract semantic relations relying on the in situ morpho-semantic structure in Chinese which can dispense of an outside source such as corpus or web data. Manual evaluation of thousands of word pairs shows that most relations can be successful predicted. We believe that it can serve as a valuable starting point in complementing with other approaches, which will hold promise for the robust lexical relations acquisition.", | |
| "pdf_parse": { | |
| "paper_id": "W14-0139", | |
| "_pdf_hash": "", | |
| "abstract": [ | |
| { | |
| "text": "Semantic relations of different types have played an important role in wordnet, and have been widely recognized in various fields. In recent years, with the growing interests of constructing semantic network in support of intelligent systems, automatic semantic relation discovery has become an urgent task. This paper aims to extract semantic relations relying on the in situ morpho-semantic structure in Chinese which can dispense of an outside source such as corpus or web data. Manual evaluation of thousands of word pairs shows that most relations can be successful predicted. We believe that it can serve as a valuable starting point in complementing with other approaches, which will hold promise for the robust lexical relations acquisition.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Abstract", | |
| "sec_num": null | |
| } | |
| ], | |
| "body_text": [ | |
| { | |
| "text": "Semantic relations are at the core of WordNetalike architecture, and constitute the essential and integral part of linguistic and conceptual knowledge formalization. However, the manual labeling task of semantic relations is very laborious.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": "1" | |
| }, | |
| { | |
| "text": "To minimize the labor, in recent years, automatic ways of extracting semantic relations from textual data have been proposed. Among these methods, extensive works have been done based on the so-called pattern-based approaches, which was pioneered by (Hearst, 1992) . The patterns predefined or plucked out of a corpus are often referred to as lexicosyntactic patterns, which serve as an information marker for a certain relation between two concepts. Later representative works using such approaches include (Cimiano et al., 2005) , and (Pantel and Pennacchiotti, 2006) , etc. Pattern-based extraction has shown quite reasonable success characterized by a (relatively) high precision rate, but suffers from a very low recall resulting from the fact that the patterns are rare in corpora. Remedies against the problem involve exploiting scaled data from the web (Cimiano et al., 2005) , but runs the risk of influenced by the web genre (Alain, 2010) .", | |
| "cite_spans": [ | |
| { | |
| "start": 250, | |
| "end": 264, | |
| "text": "(Hearst, 1992)", | |
| "ref_id": "BIBREF7" | |
| }, | |
| { | |
| "start": 508, | |
| "end": 530, | |
| "text": "(Cimiano et al., 2005)", | |
| "ref_id": "BIBREF3" | |
| }, | |
| { | |
| "start": 537, | |
| "end": 569, | |
| "text": "(Pantel and Pennacchiotti, 2006)", | |
| "ref_id": "BIBREF16" | |
| }, | |
| { | |
| "start": 861, | |
| "end": 883, | |
| "text": "(Cimiano et al., 2005)", | |
| "ref_id": "BIBREF3" | |
| }, | |
| { | |
| "start": 935, | |
| "end": 948, | |
| "text": "(Alain, 2010)", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": "1" | |
| }, | |
| { | |
| "text": "To enrich the relations coverage in Chinese Wordnet (CWN), in this paper, we propose an in situ approach by exploiting the morphsemantic information. This method, simple and straightforward as it seems, does not incur the difficulties associated with lexical gaps in cross-language mapping that any translationbased model would encounter; and it is also economic and complementary with previous approaches in that we can dispense of an outside corpus resource.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": "1" | |
| }, | |
| { | |
| "text": "In what follows, Section 2 gives a brief summary of lexical semantic relations acquisition from two perspectives. Section 3 explains the proposed methods for the automatic discovery of semantic relations, which are the main focus of this study. Section 4 shows the experiment results and discussion. Finally, we conclude this paper in Section 5.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": "1" | |
| }, | |
| { | |
| "text": "Modelling on English WordNet, CWN has been launched by Academia Sinica in 2006 and continuously broadened its scope (Huang et al., 2010 ). 1 . The initial version of CWN contains a manually created fine-grained senses repository but sparse relations. However, semantic relation labeling is a time-consuming and labor-demanding task. Two main methods were employed to automatic relation acquisition.", | |
| "cite_spans": [ | |
| { | |
| "start": 116, | |
| "end": 135, | |
| "text": "(Huang et al., 2010", | |
| "ref_id": "BIBREF10" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Relations in Chinese Wordnet", | |
| "sec_num": "2" | |
| }, | |
| { | |
| "text": "Though lexical semantic relations (LSRs) could be presumed to be more universal than word senses in human languages, a direct copying or simple porting of LSRs from one wordnet to another could possibly lead to invalid relations in the target wordnet. A broader view on the underlying inference logic of cross-language LSRs with 26 rules was first proposed by and formally introduced in (Hsieh, 2009) . A series of large-scaled bilingual bootstrapping experiments showed substantial improvements (with 55% precision) over baseline model (47%). However, it was also reported that among the correctly predicted LSRs, a large portion (c.a. 60%) belongs to non-lexical relations such as similar to, pertainym, also see, etc.", | |
| "cite_spans": [ | |
| { | |
| "start": 387, | |
| "end": 400, | |
| "text": "(Hsieh, 2009)", | |
| "ref_id": "BIBREF8" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Bilingual Bootstrapping Approach", | |
| "sec_num": "2.1" | |
| }, | |
| { | |
| "text": "To look deeper into the issues, second experiment focusing only on the hypernymytroponymy among the verbs was conducted. The bootstrapping model returned totally 12214 verb pairs mapped from WordNet 3.0, which were manually evaluated. The analysis shows that around 50% verb pairs can be recognized as fit in CWN, however, two main error types are identified: [1] Lexicalization of verbs: similar to the problems of lexical gap appeared in the cross-language sense mapping, a single word in English often has meanings that require several words in Chinese to explain. By analyzing the results, it is found that many verbs could not be described by a single lexeme in Chinese.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Bilingual Bootstrapping Approach", | |
| "sec_num": "2.1" | |
| }, | |
| { | |
| "text": "[2] Mismatch of synset: other than the above, there are cases when the hypernymy-troponymy relations of the verb pairs are approved, but the synset that CWN chooses is not the same with that of PWN. This could be due to the different semantic ranges between CWN and PWN hypernymytroponymy pairs, or due to the subtlety of sense division when the sense levels are similar.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Bilingual Bootstrapping Approach", | |
| "sec_num": "2.1" | |
| }, | |
| { | |
| "text": "The bilingual bootstrapping experiments showed that lexical relations turn out to be not subject to automatic importing and would still require tremendous human efforts of validations.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Bilingual Bootstrapping Approach", | |
| "sec_num": "2.1" | |
| }, | |
| { | |
| "text": "There has been a variety of studies on the automatic acquisition of lexical semantic relations, Hearst (Hearst, 1992) first proposed a lexico-syntactic pattern based method for automatic acquisition of hyponymy from unrestricted texts, and since then automatically finding semantic relations by using various pattern-based algorithm has become the most common approach.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Pattern-based Approach", | |
| "sec_num": "2.2" | |
| }, | |
| { | |
| "text": "We (Lo et al., 2008) have tried to define some patterns (e.g., a manner of ) to extract troponymy among verbs in Chinese. To avoid the interference of unnecessary contextual information which may include modal verbs, hedging, negation that often occur in different corpus genres, we applied the proposed patterns on the gloss of CWN. The results were evaluated with the substitution tests. Substitution test is commonly used in linguistic literature ; EuroWordnet provided linguistic tests for each semantic relation to examine the validity. In , sentence formulae were created following the frame in EuroWordnet to examine the validity of certain semantic relations in Chinese. Linguistic semantic tests help researcher check if two word meanings have a certain kind of semantic relation or not, and further ensure the quality and consistence of the database. Therefore, following the previous framework, a set of sentence formulae based on properties of troponymy was created to verify the correctness of hypernymy-troponymy verb pairs. However, due to data sparseness, the system can achieve only high precision but low recall.", | |
| "cite_spans": [ | |
| { | |
| "start": 3, | |
| "end": 20, | |
| "text": "(Lo et al., 2008)", | |
| "ref_id": "BIBREF13" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Pattern-based Approach", | |
| "sec_num": "2.2" | |
| }, | |
| { | |
| "text": "Instead of assuming any external context in which words to be linked appear, we propose to exploit the language-internal evidence manifested at the morpho-syntactic levels in Chinese, which is assumably guided by underlying semantic composition of morphemes.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Morpho-semantic Linkage", | |
| "sec_num": "3" | |
| }, | |
| { | |
| "text": "The idea of exploiting morpho-semantic information for the enrichment of WordNet has been discussed and implemented in the Word-Net community for a while. (Miller and Fellbaum, 2003) first described the importance of adding \"morphosemantic links\" to WordNet, with later works (Fellbaum et al., 2009) on the classification of regular polysemous patterns of morphosemantic V-N pairs related via -er affixation (e.g., build-builder).", | |
| "cite_spans": [ | |
| { | |
| "start": 155, | |
| "end": 182, | |
| "text": "(Miller and Fellbaum, 2003)", | |
| "ref_id": "BIBREF14" | |
| }, | |
| { | |
| "start": 276, | |
| "end": 299, | |
| "text": "(Fellbaum et al., 2009)", | |
| "ref_id": "BIBREF5" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Morpho-semantics in WordNets", | |
| "sec_num": "3.1" | |
| }, | |
| { | |
| "text": "The notion of morpho-semantic links (MSLs) has been applied to other (morphologically-rich) languages such as Czech (Pala and Hlav\u00e1ckov\u00e1, 2007) (in terms of D-relations), Turkish (Bilgin et al., 2004) and Bantu languages (Bosch et al., 2008) . It is worth of mentioning that the proposed morpho-semantic relations or derivational relations are relations that hold among literals (lemmas) rather than synsets, which leaves some room of discussion about the extra level these relations should be anchored because neither paradigmatic nor syntagmatic relations would fit.", | |
| "cite_spans": [ | |
| { | |
| "start": 116, | |
| "end": 143, | |
| "text": "(Pala and Hlav\u00e1ckov\u00e1, 2007)", | |
| "ref_id": "BIBREF15" | |
| }, | |
| { | |
| "start": 179, | |
| "end": 200, | |
| "text": "(Bilgin et al., 2004)", | |
| "ref_id": "BIBREF1" | |
| }, | |
| { | |
| "start": 221, | |
| "end": 241, | |
| "text": "(Bosch et al., 2008)", | |
| "ref_id": "BIBREF2" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Morpho-semantics in WordNets", | |
| "sec_num": "3.1" | |
| }, | |
| { | |
| "text": "It is note here that for morphologically-poor languages like Chinese, the MSLs are quite different in that they do not exist between stems and suffixes, but between word-to-be/wordused-to-be morphemes instead. This has the practical advantages for the enrichment of existing paradigmatic relations, as we will introduce in the following.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Morpho-semantics in WordNets", | |
| "sec_num": "3.1" | |
| }, | |
| { | |
| "text": "The vast majority of Chinese characters represent the morphemes. It has been always a controversy over the notion of wordhood in the lexical history of Chinese. In a way any Chinese character can be seen as word-to-be or word-used-to-be morphemes. Given the fact that the relative predominance of the monosyllabic word in ancient Chinese has shifted to bi-syllabic words in modern Chinese, the huge semantic weight carried by the morphemes has made the idea of character-centered lexicon deeply ingrained in Chinese mind. Orthographically, the lack of word delimiter (such as space) in texts worsens the achievement of consensus regarding the distinction between words, compounds and phrases, and thus makes the segmentation a long-standing heated topic in Chinese NLP. We follow the cognitive-functional stance in the respect that lexicon and syntax form a continuum rather than two strictly separated modules. We argue that the Morpho-Semantic Relations (MSRs), i.e., the ways morphemes combine to form composite meanings, can function as the organic linkage in revealing the composition mechanism among the continuum of different lexical units in varied contexts. In terms of WordNet's paradigmatic relations, this means that morpho-semantic in-formation in Chinese can be used to identify these relations based on the position and semantic role of morphemes in modification.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Probing Morpho-Semantic Relations in Chinese", | |
| "sec_num": "3.2" | |
| }, | |
| { | |
| "text": "In the case of Verb-Verb (compound) words, where the word is composed of two verbal morphemes, linguistics have sorted out different types resulting from the interplay of morphemes within (Li and Thompson, 1981) . For instance, for the type of so-called 'parallel' VV compounds, V 1 (verb in the first position) and V 2 (verb in the second position) share the similar meaning (near synonyms), such as bang-zhu 'help-assist' (help), fang-qi 'loosenabandon' (give up). With a fine-grained sense analysis, we can label the troponymy between V 1 and V 1 V 2 , where V 1 is widely recognized as the component that carries heavier semantic load in VV compound (a.k.a. left-headedness).", | |
| "cite_spans": [ | |
| { | |
| "start": 188, | |
| "end": 211, | |
| "text": "(Li and Thompson, 1981)", | |
| "ref_id": "BIBREF12" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Probing Morpho-Semantic Relations in Chinese", | |
| "sec_num": "3.2" | |
| }, | |
| { | |
| "text": "In the case of Noun-Noun (compound) words, e.g., noodle-shop 'mian-dian' (noodle shop), where the word is composed of two nominal morphemes, the N 1 modifier -N 2 head structure is prevalently observed (a.k.a. rightheadedness). The linkage between N 1 N 2 and N 2 can be labeled as hypernymy-hyponymy.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Probing Morpho-Semantic Relations in Chinese", | |
| "sec_num": "3.2" | |
| }, | |
| { | |
| "text": "As argued in previous section, Morpho-Semantic Linkage abound in abundant relational knowledge. In this study, we aim to enrich the CWN with relations leveraged by operationalizing MSL. The automatic labeling of the lexical semantic relations on word-pairs is quite straightforward. For N 1 N 2 compounds, \u227a N 1 N 2 , N 2 \u227b pairs are labeled with hypernymy-hyponymy, and \u227aN 1 N 2 , N 1 \u227b pairs are labeled with meronymy-holonymy.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Hypothesis", | |
| "sec_num": "4.1" | |
| }, | |
| { | |
| "text": "The cases of VV compounds are trickier, the flow of judgement is shown in algorithm 1. When V 1 has synonymy or near-synonymy ", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Hypothesis", | |
| "sec_num": "4.1" | |
| }, | |
| { | |
| "text": "with V 2 , then V 1 V 2 are troponyms of both V 1 and V 2 . If V", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Hypothesis", | |
| "sec_num": "4.1" | |
| }, | |
| { | |
| "text": "V 1 V 2 and V 1 /V 2", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Hypothesis", | |
| "sec_num": "4.1" | |
| }, | |
| { | |
| "text": "In this section, we discuss the experiment we designed, the evaluation and error analysis.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Experiments", | |
| "sec_num": "4.2" | |
| }, | |
| { | |
| "text": "The first step is to create a list of term pairs, which a total of 561,703 words covered in CWN 2 , Sinica BOW 3 , and Ministry of Education Online Chinese Dictionary 4 . In this experiment, we focus only on bi-syllabic words represented by two characters, which constitute the largest proportion of Chinese vocabulary repository.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Experiments", | |
| "sec_num": "4.2" | |
| }, | |
| { | |
| "text": "In order to filter out a coarse-grained bisyllabic word list, only both characters of a bi-syllabic word that could be found in the big word list, are preserved. Additionally, four principles are applied to construct a more finegrained word list: [1] the part-of-speech tags of both characters within a bi-syllabic word should be NN or VV; [2] bi-syllabic words containing metaphors are excluded; [3] bi-syllabic morphemic word (e.g., \u9f77\u9f6a (sordid)) or archaic words (e.g.,\u6417\u5bb6) are not included; and [4] proper nouns, (e.g., \u6210\u9f8d(Jackie Chan)) are not considered. Therefore, a list with 1482 bisyllabic words are produced. Using the hypotheses proposed in section 4.1, the relations are automatically labelled on the related word pairs.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Experiments", | |
| "sec_num": "4.2" | |
| }, | |
| { | |
| "text": "A manual evaluation of the resulting semantic relations lists was conducted. We have created a wiki-based collaborative platform 5 on which registered users can contribute to CWN by adding new entries, editing existing ones and rating one another's contribution to ensure the quality of collective intelligence (Lee et al, 2013) . Figure 1 shows the snapshot of the system.", | |
| "cite_spans": [ | |
| { | |
| "start": 311, | |
| "end": 328, | |
| "text": "(Lee et al, 2013)", | |
| "ref_id": "BIBREF11" | |
| } | |
| ], | |
| "ref_spans": [ | |
| { | |
| "start": 331, | |
| "end": 339, | |
| "text": "Figure 1", | |
| "ref_id": "FIGREF0" | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Experiments", | |
| "sec_num": "4.2" | |
| }, | |
| { | |
| "text": "With three linguistic graduate students judging the correctness, the inter-annotator agreement measured by Fleiss kappa (Fleiss, 1971) was used, which is defined as: k = P \u2212 P e 1 \u2212 P e where the numerator expresses the degree of agreement actually achieved, and the denominator the degree of agreement that is possible above chance. As a result, it's interesting to see that there is a very poor agreement between three raters (k = \u22120.7069972) on the predicted relations of \u227a W 1 \u2212 W 1 W 2 \u227b, which also gets low precision rate; while agreement achieves a moderate degree (with k = 0.5835113) on the predicted relations of \u227a W 2 \u2212 W 1 W 2 \u227b, which also gets high performance in precision. 6 Figure 2 shows the enrichment of relations through the experiment.", | |
| "cite_spans": [ | |
| { | |
| "start": 120, | |
| "end": 134, | |
| "text": "(Fleiss, 1971)", | |
| "ref_id": "BIBREF6" | |
| } | |
| ], | |
| "ref_spans": [ | |
| { | |
| "start": 692, | |
| "end": 700, | |
| "text": "Figure 2", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Experiments", | |
| "sec_num": "4.2" | |
| }, | |
| { | |
| "text": "The experiment we carried out gives rise to some issues for discussion. Table 1 shows the performance for each predicted relation. When we scrutinize the portion with low precision rate, we found that the problematic cases are mostly from the predicted meronymyholonymy relations between NN compounds, i.e., \u227a N 1 N 2 \u2212 N 1 \u227b. It is in fact not surprising in that the definition of part-whole is not easily stated, and the judgement criteria in the previous literature are not unproblematic too. For instance, given the restrictive rules that Cruse (1986) sets on the meronymy relation with the co-existence of both the 'N 1 N 2 is part of N 1 ' and 'N 1 has N 1 N 2 ' paraphrases, the raters did not all agree that the relation hold between \u9ee8\u90e8 (party headquarter) and \u9ee8 (political party).", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 72, | |
| "end": 79, | |
| "text": "Table 1", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "Discussion", | |
| "sec_num": "4.3" | |
| }, | |
| { | |
| "text": "Another main error sources come from the predicted troponymy-hypernymy relations between \u227a", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Discussion", | |
| "sec_num": "4.3" | |
| }, | |
| { | |
| "text": "V 1 V 2 \u2212 V 1 \u227b.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Discussion", | |
| "sec_num": "4.3" | |
| }, | |
| { | |
| "text": "Recall that we hypothesize that if V 1 and V 2 are synonymous, then V 1 V 2 is automatically labeled as troponym of V 1 . The errors arose here can be mainly ascribed to the lack of consistent Chinese thesaurus. In this experiment, the CWN synset (fine-grained synonym determination) and CILIN semantic class (coarse-grained synonym determination) are integrated for prediction, both has different criteria regarding the sameness or nearness of senses between two verbs. In addition, no proper rules for the evaluation of troponymy among raters constitute the difficulties as well.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Discussion", | |
| "sec_num": "4.3" | |
| }, | |
| { | |
| "text": "Furthermore, there are two points can be made.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Discussion", | |
| "sec_num": "4.3" | |
| }, | |
| { | |
| "text": "[1], the experiment of relation discovery is conducted at the level of word-lemma, not concept(word-sense), in terms of wordnet, the generic label 'semantic relations' are regarded as the relation occurring between linguistic units rather than between concepts (i.e., synsets.) Currently, the predicted relations are presumably connected with the first sense of the word lemma in CWN. A fine-grained annotation will be left for future work.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Discussion", | |
| "sec_num": "4.3" | |
| }, | |
| { | |
| "text": "[2], in the evaluation task, when the raters did not agree with the predicted relation type, they also provide proper relation types for the pair, which are not named relations explicitly defined in WordNet. For example, the qualia modification between certain N 1 N 2 and N 2 , such as \u8089\u91ac(meat sauce) -\u91ac(sauce). This is different from patternedbased approaches where a bottom-up methodology is taken because named and explicitly defined semantic relations of interest are presumed before lexico-syntactic patterns are extracted and utilized to search for instances of the relations", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Discussion", | |
| "sec_num": "4.3" | |
| }, | |
| { | |
| "text": "Lexical semantic relations offers rich linguistic and conceptual knowledge information and are the most to fill in for wordnets. Semantic relations extraction has been one of the most important tasks in many fields. The challenges pertaining to this task are multifaceted. The most active pattern-based approaches provide a reasonable solution, but poses difficulties as well.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Conclusion", | |
| "sec_num": "5" | |
| }, | |
| { | |
| "text": "In this paper, we have presented a linguistic alternative to the task in Chinese by resorting to resources of language in itself. Rather than focusing on the patterns design -relation extraction model, a notion of Morpo-semantic links is proposed to support the extraction and labeling of a wide variety of semantic relations in Chinese. The experiment shows that it is possible to discover semantic relations without being influenced by corpus size and genres. This simple strategy can also serve as the linguistic baseline for related works.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Conclusion", | |
| "sec_num": "5" | |
| }, | |
| { | |
| "text": "Future works include: [1] extending to VN and NV compounds (Song and Qiu, 1981) , and more fined-grained classification of semantic relations among these word-pairs, and [2] mapping with Japanese Wordnet where an amount of Chinese characters are employed for advanced cross-linguistic validation. We also hope that the work presented here will shed new light on the understanding of morphosemantic representation of natural languages.", | |
| "cite_spans": [ | |
| { | |
| "start": 59, | |
| "end": 79, | |
| "text": "(Song and Qiu, 1981)", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Conclusion", | |
| "sec_num": "5" | |
| }, | |
| { | |
| "text": "Freely available at http://lope.linguistics. ntu.edu.tw/cwn", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "See http://lope.linguistics.ntu.edu.tw/cwn/ 3 See http://bow.sinica.edu.tw/ 4 See http://dict.revised.moe.edu.tw/", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "See http://lope.linguistics.ntu.edu.tw/ cwikin/ 6 The results will be accessible at http://140.112. 147.131/", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| } | |
| ], | |
| "back_matter": [], | |
| "bib_entries": { | |
| "BIBREF0": { | |
| "ref_id": "b0", | |
| "title": "Probing Semantic Relations", | |
| "authors": [], | |
| "year": 2010, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Alain Auger and Caroline Barri\u00e8re (eds). 2010. Probing Semantic Relations. John Benjamins Publishing, Amsterdam/Philadelphia.", | |
| "links": null | |
| }, | |
| "BIBREF1": { | |
| "ref_id": "b1", | |
| "title": "Morphosemantic Relations In and Across Wordnets: A Study Based on Turkish", | |
| "authors": [ | |
| { | |
| "first": "Orhan", | |
| "middle": [], | |
| "last": "Bilgin", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "\u00d6zlem", | |
| "middle": [], | |
| "last": "\u00c7etinoglu", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Kemal", | |
| "middle": [], | |
| "last": "Oflazer", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2004, | |
| "venue": "Proceedings of the Second Global WordNet Conference", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "60--66", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Orhan Bilgin, \u00d6zlem \u00c7etinoglu and Kemal Oflazer. 2004. Morphosemantic Relations In and Across Wordnets: A Study Based on Turkish. In: Pro- ceedings of the Second Global WordNet Confer- ence, 60-66.", | |
| "links": null | |
| }, | |
| "BIBREF2": { | |
| "ref_id": "b2", | |
| "title": "Enhancing WordNets with Morphological Relations: A Case Study from Czech, English and Zulu", | |
| "authors": [ | |
| { | |
| "first": "Sonja", | |
| "middle": [], | |
| "last": "Bosch", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Christiane", | |
| "middle": [], | |
| "last": "Fellbaum", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Karel", | |
| "middle": [], | |
| "last": "Pala", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2008, | |
| "venue": "Proceedings of the Fourth Global WordNet Conference", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "74--90", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Sonja Bosch, Christiane Fellbaum and Karel Pala. 2008. Enhancing WordNets with Morphological Relations: A Case Study from Czech, English and Zulu. In: Proceedings of the Fourth Global WordNet Conference, 74-90.", | |
| "links": null | |
| }, | |
| "BIBREF3": { | |
| "ref_id": "b3", | |
| "title": "Learning Taxonomic Relations from Heterogeneous Sources of", | |
| "authors": [ | |
| { | |
| "first": "Philipp", | |
| "middle": [], | |
| "last": "Cimiano", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Aleksander Pivk", | |
| "middle": [], | |
| "last": "Lars", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Steffen", | |
| "middle": [], | |
| "last": "Schmidt-Thieme", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "", | |
| "middle": [], | |
| "last": "Staab", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2005, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Philipp Cimiano, Aleksander Pivk Lars, Schmidt- Thieme and Steffen Staab. 2005. Learning Tax- onomic Relations from Heterogeneous Sources of", | |
| "links": null | |
| }, | |
| "BIBREF4": { | |
| "ref_id": "b4", | |
| "title": "Table 1: Inter-annotator agreement across all relations Evidence", | |
| "authors": [], | |
| "year": null, | |
| "venue": "Ontology Learning from Text: Methods, Evaluation and Applications", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "55--73", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Table 1: Inter-annotator agreement across all relations Evidence. In: Buitelaar (eds),Ontology Learn- ing from Text: Methods, Evaluation and Appli- cations, 55-73.", | |
| "links": null | |
| }, | |
| "BIBREF5": { | |
| "ref_id": "b5", | |
| "title": "Putting Semantics into Word-Net's \"Morphosemantic\" Links", | |
| "authors": [ | |
| { | |
| "first": "Christiane", | |
| "middle": [], | |
| "last": "Fellbaum", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Anne", | |
| "middle": [], | |
| "last": "Osherson", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Peter", | |
| "middle": [ | |
| "E" | |
| ], | |
| "last": "Clark", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2009, | |
| "venue": "Human Language Technology. Challenges of the Information Society", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "350--358", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Christiane Fellbaum, Anne Osherson, and Peter. E. Clark. 2009. Putting Semantics into Word- Net's \"Morphosemantic\" Links. In: Zygmunt Vetulani and Hans Uszkoreit (eds). Human Lan- guage Technology. Challenges of the Information Society, 350-358.", | |
| "links": null | |
| }, | |
| "BIBREF6": { | |
| "ref_id": "b6", | |
| "title": "Measuring Nominal Scale Agreement among Many Raters", | |
| "authors": [ | |
| { | |
| "first": "Joseph", | |
| "middle": [ | |
| "L" | |
| ], | |
| "last": "Fleiss", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1971, | |
| "venue": "Psychological Bulletin", | |
| "volume": "76", | |
| "issue": "5", | |
| "pages": "378--382", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Joseph. L. Fleiss. 1971. Measuring Nominal Scale Agreement among Many Raters. Psychological Bulletin, 76(5): 378-382.", | |
| "links": null | |
| }, | |
| "BIBREF7": { | |
| "ref_id": "b7", | |
| "title": "Automatic Acquisition of Hyponyms from Large Text Corpora", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Marti", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "", | |
| "middle": [], | |
| "last": "Hearst", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1992, | |
| "venue": "Proceedings of the Fourth International Conference on Computational Linguistics (COLING)", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "539--545", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Marti A. Hearst. 1992. Automatic Acquisition of Hyponyms from Large Text Corpora. In: Pro- ceedings of the Fourth International Conference on Computational Linguistics (COLING), 539- 545.", | |
| "links": null | |
| }, | |
| "BIBREF8": { | |
| "ref_id": "b8", | |
| "title": "Formal Description of Lexical Semantic Relations. Concentric: Studies in Linguistics", | |
| "authors": [ | |
| { | |
| "first": "Shu-Kai", | |
| "middle": [], | |
| "last": "Hsieh", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2009, | |
| "venue": "", | |
| "volume": "35", | |
| "issue": "", | |
| "pages": "87--109", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Shu-Kai Hsieh. 2009. Formal Description of Lexi- cal Semantic Relations. Concentric: Studies in Linguistics, 35(1):87-109.", | |
| "links": null | |
| }, | |
| "BIBREF9": { | |
| "ref_id": "b9", | |
| "title": "Translating Lexical Semantic Relations", | |
| "authors": [ | |
| { | |
| "first": "Chu-Ren", | |
| "middle": [], | |
| "last": "Huang", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "I-Ju", | |
| "middle": [], | |
| "last": "Tseng", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Dylan", | |
| "middle": [], | |
| "last": "Tsai", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2002, | |
| "venue": "SEMANET '02 Proceedings of the 2002 workshop on Building and using Semantic Networks", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "1--7", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Chu-Ren Huang, I-Ju Tseng, and Dylan Tsai. 2002. Translating Lexical Semantic Relations. In: SEMANET '02 Proceedings of the 2002 workshop on Building and using Semantic Net- works, 1-7.", | |
| "links": null | |
| }, | |
| "BIBREF10": { | |
| "ref_id": "b10", | |
| "title": "Chinese Wordnet: Design, Implementation, and Application of an Infrastructure for Cross-Lingual Knowledge Processing", | |
| "authors": [ | |
| { | |
| "first": "Chu-Ren", | |
| "middle": [], | |
| "last": "Huang", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Shu-Kai", | |
| "middle": [], | |
| "last": "Hsieh", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Jia-Fei", | |
| "middle": [], | |
| "last": "Hong", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Yun-Zhu", | |
| "middle": [], | |
| "last": "Chen", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "I-Li", | |
| "middle": [], | |
| "last": "Su", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Yong-Xiang", | |
| "middle": [], | |
| "last": "Chen", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Sheng-Wei", | |
| "middle": [], | |
| "last": "Huang", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2010, | |
| "venue": "Journal of Chinese Information Processing", | |
| "volume": "24", | |
| "issue": "2", | |
| "pages": "14--23", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Chu-Ren Huang,Shu-Kai Hsieh,Jia-Fei Hong,Yun- Zhu Chen,I-Li Su,Yong-Xiang Chen and Sheng- Wei Huang. 2010. Chinese Wordnet: Design, Implementation, and Application of an Infras- tructure for Cross-Lingual Knowledge Process- ing. Journal of Chinese Information Processing, 24(2):14-23.", | |
| "links": null | |
| }, | |
| "BIBREF11": { | |
| "ref_id": "b11", | |
| "title": "CWIKIN: A wiki that Helps Quicken the Development of Chinese Wordnet. The 8th International Conference of the Asian Association for Lexicography", | |
| "authors": [ | |
| { | |
| "first": "Chih-Yao", | |
| "middle": [], | |
| "last": "Lee", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Yu-Yun", | |
| "middle": [], | |
| "last": "Chang", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Shu-Kai", | |
| "middle": [], | |
| "last": "Hsieh", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Jia-Fei", | |
| "middle": [], | |
| "last": "Hong", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Chu-Ren", | |
| "middle": [], | |
| "last": "Huang", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2013, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Chih-Yao Lee, Yu-Yun Chang, Shu-Kai Hsieh, Jia- Fei Hong and Chu-Ren Huang. 2013. CWIKIN: A wiki that Helps Quicken the Development of Chinese Wordnet. The 8th International Confer- ence of the Asian Association for Lexicography. Bali, Indonesia.", | |
| "links": null | |
| }, | |
| "BIBREF12": { | |
| "ref_id": "b12", | |
| "title": "Mandarin Chinese: A Functional Reference Grammar", | |
| "authors": [ | |
| { | |
| "first": "Charles", | |
| "middle": [ | |
| "N" | |
| ], | |
| "last": "Li", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Sandra", | |
| "middle": [ | |
| "A" | |
| ], | |
| "last": "Thompson", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1981, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Charles N. Li and Sandra A. Thompson. 1981. Mandarin Chinese: A Functional Reference Grammar. Berkeley: University of California Press.", | |
| "links": null | |
| }, | |
| "BIBREF13": { | |
| "ref_id": "b13", | |
| "title": "Automatic Labeling of Troponymy for Chinese Verbs", | |
| "authors": [ | |
| { | |
| "first": "Chiao-Shan", | |
| "middle": [], | |
| "last": "Lo", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Yi-Rung", | |
| "middle": [], | |
| "last": "Chen", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Chih-Yu", | |
| "middle": [], | |
| "last": "Lin", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Shu-Kai", | |
| "middle": [], | |
| "last": "Hsieh", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2008, | |
| "venue": "Proceedings of the 20th Conference on Computational Linguistics and Speech Processing", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Chiao-Shan Lo, Yi-Rung Chen, Chih-Yu Lin, and Shu-Kai Hsieh. 2008. Automatic Labeling of Troponymy for Chinese Verbs. In: Proceedings of the 20th Conference on Computational Lin- guistics and Speech Processing (ROCLING).", | |
| "links": null | |
| }, | |
| "BIBREF14": { | |
| "ref_id": "b14", | |
| "title": "Morphosemantic Links in WordNet", | |
| "authors": [ | |
| { | |
| "first": "George", | |
| "middle": [], | |
| "last": "Miller", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Christiane", | |
| "middle": [], | |
| "last": "Fellbaum", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2003, | |
| "venue": "Traitement Automatique de Langue", | |
| "volume": "44", | |
| "issue": "2", | |
| "pages": "69--80", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "George Miller and Christiane Fellbaum. 2003. Morphosemantic Links in WordNet. Traitement Automatique de Langue, 44(2):69-80.", | |
| "links": null | |
| }, | |
| "BIBREF15": { | |
| "ref_id": "b15", | |
| "title": "Derivational Relations in Czech WordNet", | |
| "authors": [ | |
| { | |
| "first": "Karel", | |
| "middle": [], | |
| "last": "Pala", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Dana", | |
| "middle": [], | |
| "last": "Hlav\u00e1ckov\u00e1", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2007, | |
| "venue": "ACL '07 Proceedings of the Workshop on Balto-Slavonic Natural Language Processing: Information Extraction and Enabling Technologies", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Karel Pala and Dana Hlav\u00e1ckov\u00e1. 2007. Deriva- tional Relations in Czech WordNet. In: ACL '07 Proceedings of the Workshop on Balto-Slavonic Natural Language Processing: Information Ex- traction and Enabling Technologies. Strouds- burg, PA, USA.", | |
| "links": null | |
| }, | |
| "BIBREF16": { | |
| "ref_id": "b16", | |
| "title": "Espresso: Leveraging Generic Patterns for Automatically Harvesting Semantic Relations", | |
| "authors": [ | |
| { | |
| "first": "Patrick", | |
| "middle": [], | |
| "last": "Pantel", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Marco", | |
| "middle": [], | |
| "last": "Pennacchiotti", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2006, | |
| "venue": "Proceedings of Conference on Computational Linguistics / Association for Computational Linguistics (COLING/ACL-06)", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Patrick Pantel and Marco Pennacchiotti. 2006. Espresso: Leveraging Generic Patterns for Au- tomatically Harvesting Semantic Relations. In: Proceedings of Conference on Computational Linguistics / Association for Computational Linguistics (COLING/ACL-06).", | |
| "links": null | |
| }, | |
| "BIBREF17": { | |
| "ref_id": "b17", | |
| "title": "Definitional Verbal Patterns for Semantic Relation Extraction", | |
| "authors": [ | |
| { | |
| "first": "Gerardo", | |
| "middle": [], | |
| "last": "Sierra", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Rodrigo", | |
| "middle": [], | |
| "last": "Alarc\u00f3n", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "C\u00e9sar", | |
| "middle": [], | |
| "last": "Aguilar", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Carme", | |
| "middle": [], | |
| "last": "Bach", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2010, | |
| "venue": "Probing Semantic Relations. John Benjamins Publishing", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Gerardo Sierra, Rodrigo Alarc\u00f3n, C\u00e9sar Aguilar and Carme Bach. 2010. Definitional Verbal Patterns for Semantic Relation Extraction. In: Alain Auger and Caroline Barri\u00e8re (eds). Prob- ing Semantic Relations. John Benjamins Pub- lishing, Amsterdam/Philadelphia.", | |
| "links": null | |
| }, | |
| "BIBREF18": { | |
| "ref_id": "b18", | |
| "title": "Qualia Relations in Chinese Nominal Compounds Containing Verbal Elements", | |
| "authors": [ | |
| { | |
| "first": "Zuoyan", | |
| "middle": [], | |
| "last": "Song", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Qiu", | |
| "middle": [], | |
| "last": "Likun", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2013, | |
| "venue": "International Journal of Knowledge and Language Processing", | |
| "volume": "28", | |
| "issue": "1", | |
| "pages": "114--133", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Zuoyan Song and Qiu Likun. 2013. Qualia Rela- tions in Chinese Nominal Compounds Contain- ing Verbal Elements. International Journal of Knowledge and Language Processing, 28(1):114- 133.", | |
| "links": null | |
| }, | |
| "BIBREF19": { | |
| "ref_id": "b19", | |
| "title": "Chinese Lexical Semantic Relations: Definition and Classification Criteria", | |
| "authors": [ | |
| { | |
| "first": "Dylan", | |
| "middle": [ | |
| "B" | |
| ], | |
| "last": "Tsai", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Chu-Ren", | |
| "middle": [], | |
| "last": "Huang", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Shu-Chuan", | |
| "middle": [], | |
| "last": "Tseng", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "J", | |
| "middle": [ | |
| "I" | |
| ], | |
| "last": "Elanna", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "", | |
| "middle": [], | |
| "last": "Lin", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 2002, | |
| "venue": "Journal of Chinese Information Processing", | |
| "volume": "", | |
| "issue": "4", | |
| "pages": "21--31", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Dylan B. Tsai, Chu-Ren Huang, Shu-Chuan Tseng, Elanna J.I.Lin, Keh-jiann Chen and Yuan-hsun Chuang. 2002. Chinese Lexical Semantic Relations: Definition and Classification Crite- ria. Journal of Chinese Information Processing, 2002(4):21-31.", | |
| "links": null | |
| } | |
| }, | |
| "ref_entries": { | |
| "FIGREF0": { | |
| "uris": null, | |
| "type_str": "figure", | |
| "text": "User graphical interface of CWIKIN Figure 2: Relations added", | |
| "num": null | |
| }, | |
| "TABREF0": { | |
| "type_str": "table", | |
| "text": "2 is on the list of \u5b8c\u4f4f\u6389\u958b\u58de \u6210, which is a subclass of the VV compounds that are often called resultative compounds, for there is a causal relation between the event represented by the first compound of such a compound and the event/state represented by the second component.", | |
| "num": null, | |
| "content": "<table><tr><td>Data: VV compounds</td></tr><tr><td>Result: Labeled relations between V 1 V 2</td></tr><tr><td>and V 1 /V 2</td></tr><tr><td>initialization (POS tagging);</td></tr><tr><td>if V1 is V2 then</td></tr><tr><td>return troponymy;</td></tr><tr><td>else</td></tr><tr><td>if V2 is \u5b8c\u4f4f\u6389\u958b\u58de\u6210 then</td></tr><tr><td>return causality;</td></tr><tr><td>if V2 is \u4e0a\u4e0b\u4f86\u53bb\u9032\u56de\u51fa\u843d\u5165\u5411\u5f80\u904e\u8d77</td></tr><tr><td>then</td></tr><tr><td>return directional;</td></tr><tr><td>end</td></tr><tr><td>else</td></tr><tr><td>return pertainymy;</td></tr><tr><td>end</td></tr><tr><td>end</td></tr><tr><td>Algorithm 1: Pseudo code for relations la-</td></tr><tr><td>beling between</td></tr></table>", | |
| "html": null | |
| } | |
| } | |
| } | |
| } |