{ "paper_id": "H01-1043", "header": { "generated_with": "S2ORC 1.0.0", "date_generated": "2023-01-19T03:31:13.291777Z" }, "title": "Japanese Case Frame Construction by Coupling the Verb and its Closest Case Component", "authors": [ { "first": "Daisuke", "middle": [], "last": "Kawahara", "suffix": "", "affiliation": { "laboratory": "", "institution": "Kyoto University Yoshida-Honmachi", "location": { "addrLine": "Sakyo-ku", "postCode": "606-8501", "settlement": "Kyoto", "country": "Japan" } }, "email": "kawahara@pine.kuee.kyoto-u.ac.jp" }, { "first": "Sadao", "middle": [], "last": "Kurohashi", "suffix": "", "affiliation": {}, "email": "" } ], "year": "", "venue": null, "identifiers": {}, "abstract": "This paper describes a method to construct a case frame dictionary automatically from a raw corpus. The main problem is how to handle the diversity of verb usages. We collect predicate-argument examples, which are distinguished by the verb and its closest case component in order to deal with verb usages, from parsed results of a corpus. Since these couples multiply to millions of combinations, it is difficult to make a wide-coverage case frame dictionary from a small corpus like an analyzed corpus. We, however, use a raw corpus, so that this problem can be addressed. Furthermore, we cluster and merge predicate-argument examples which does not have different usages but belong to different case frames because of different closest case components. We also report on an experimental result of case structure analysis using the constructed case frame dictionary.", "pdf_parse": { "paper_id": "H01-1043", "_pdf_hash": "", "abstract": [ { "text": "This paper describes a method to construct a case frame dictionary automatically from a raw corpus. The main problem is how to handle the diversity of verb usages. We collect predicate-argument examples, which are distinguished by the verb and its closest case component in order to deal with verb usages, from parsed results of a corpus. Since these couples multiply to millions of combinations, it is difficult to make a wide-coverage case frame dictionary from a small corpus like an analyzed corpus. We, however, use a raw corpus, so that this problem can be addressed. Furthermore, we cluster and merge predicate-argument examples which does not have different usages but belong to different case frames because of different closest case components. We also report on an experimental result of case structure analysis using the constructed case frame dictionary.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Abstract", "sec_num": null } ], "body_text": [ { "text": "Syntactic analysis or parsing has been a main objective in Natural Language Processing. In case of Japanese, however, syntactic analysis cannot clarify relations between words in sentences because of several troublesome characteristics of Japanese such as scrambling, omission of case components, and disappearance of case markers. Therefore, in Japanese sentence analysis, case structure analysis is an important issue, and a case frame dictionary is necessary for the analysis.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "INTRODUCTION", "sec_num": "1." }, { "text": "Some research institutes have constructed Japanese case frame dictionaries manually [2, 3] . However, it is quite expensive, or almost impossible to construct a wide-coverage case frame dictionary by hand.", "cite_spans": [ { "start": 84, "end": 87, "text": "[2,", "ref_id": null }, { "start": 88, "end": 90, "text": "3]", "ref_id": "BIBREF2" } ], "ref_spans": [], "eq_spans": [], "section": "INTRODUCTION", "sec_num": "1." }, { "text": "Others have tried to construct a case frame dictionary automatically from analyzed corpora. However, existing syntactically analyzed corpora are too small to learn a dictionary, since case frame information consists of relations between nouns and verbs, which multiplies to millions of combinations. Based on such a consideration, we took the unsupervised learning strategy to Japanese case frame construction 1 .", "cite_spans": [ { "start": 410, "end": 411, "text": "1", "ref_id": "BIBREF0" } ], "ref_spans": [], "eq_spans": [], "section": "INTRODUCTION", "sec_num": "1." }, { "text": "To construct a case frame dictionary from a raw corpus, we parse a raw corpus first, but parse errors are problematic in this case. However, if we use only reliable modifier-head relations to construct a case frame dictionary, this problem can be addressed. Verb sense ambiguity is rather problematic. Since verbs can have different cases and case components depending on their meanings, verbs which have different meanings should have different case frames. To deal with this problem, we collect predicate-argument examples, which are distinguished by the verb and its closest case component, and cluster them. That is, examples are not distinguished by verbs such as naru 'make, become' and tsumu 'load, accumulate', but by couples such as tomodachi ni naru 'make a friend', byouki ni naru 'become sick',nimotsu wo tsumu 'load baggage', and keiken wo tsumu 'accumulate experience'. Since these couples multiply to millions of combinations, it is difficult to make a wide-coverage case frame dictionary from a small corpus like an analyzed corpus. We, however, use a raw corpus, so that this problem can be addressed. The clustering process is to merge examples which does not have different usages but belong to different case frames because of different closest case components.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "INTRODUCTION", "sec_num": "1." }, { "text": "We employ the following procedure of case frame construction from raw corpus ( Figure 1 ): 1. A large raw corpus is parsed by KNP [5] , and reliable modifier-head relations are extracted from the parse results. We call these modifier-head relations examples.", "cite_spans": [ { "start": 130, "end": 133, "text": "[5]", "ref_id": "BIBREF4" } ], "ref_spans": [ { "start": 79, "end": 87, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "VARIOUS METHODS FOR CASE FRAME CONSTRUCTION", "sec_num": "2." }, { "text": "The extracted examples are distinguished by the verb and its closest case component. We call these data example patterns.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "2.", "sec_num": null }, { "text": "3. The example patterns are clustered based on a thesaurus. We call the output of this process example case frames, which is the final result of the system. We call words which compose case components case examples, and a group of case examples case example group. In Figure 1 , nimotsu 'baggage', busshi 1 In English, several unsupervised methods have been proposed [7, 1] . However, it is different from those that combinations of nouns and verbs must be collected in Japanese. 'supply', and keiken 'experience' are case examples, and {nimotsu 'baggage', busshi 'supply'} (of wo case marker in the first example case frame of tsumu 'load, accumulate') is a case example group. A case component therefore consists of a case example and a case marker (CM).", "cite_spans": [ { "start": 305, "end": 306, "text": "1", "ref_id": "BIBREF0" }, { "start": 367, "end": 370, "text": "[7,", "ref_id": "BIBREF6" }, { "start": 371, "end": 373, "text": "1]", "ref_id": "BIBREF0" } ], "ref_spans": [ { "start": 268, "end": 276, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "2.", "sec_num": null }, { "text": "Let us now discuss several methods of case frame construction as shown in Figure 1 .", "cite_spans": [], "ref_spans": [ { "start": 74, "end": 82, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "2.", "sec_num": null }, { "text": "First, examples (I of Figure 1 ) can be used individually, but this method cannot solve the sparse data problem. For example, even if these two examples occur in a corpus, it cannot be judged whether the expression \"kuruma ni busshi wo tsumu\" (load supply onto the car) is allowed or not.", "cite_spans": [], "ref_spans": [ { "start": 22, "end": 30, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "2.", "sec_num": null }, { "text": "Secondly, examples can be decomposed into binomial relations (II of Figure 1 ). These co-occurrences are utilized by statistical parsers, and can address the sparse data problem. In this case, however, verb sense ambiguity becomes a serious problem. For example, from these two examples, three co-occurrences (\"kuruma ni tsumu\", \"nimotsu wo tsumu\", and \"keiken wo tsumu\") are extracted. They, however, allow the incorrect expression \"kuruma ni keiken wo tsumu\" (load experience onto the car, accumulate experience onto the car).", "cite_spans": [], "ref_spans": [ { "start": 68, "end": 76, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "2.", "sec_num": null }, { "text": "Thirdly, examples can be simply merged into one frame (III of Figure 1 ). However, information quantity of this is equivalent to that of the co-occurrences (II of Figure 1 ), so verb sense ambiguity becomes a problem as well.", "cite_spans": [], "ref_spans": [ { "start": 62, "end": 70, "text": "Figure 1", "ref_id": "FIGREF0" }, { "start": 163, "end": 171, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "2.", "sec_num": null }, { "text": "We distinguish examples by the verb and its closest case component. Our method can address the two problems above: verb sense ambiguity and sparse data.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "2.", "sec_num": null }, { "text": "On the other hand, semantic markers can be used as case components instead of case examples. These we call semantic case frames (IV of Figure 1 ). Constructing semantic case frames by hand leads to the problem mentioned in Section 1. Utsuro et al. constructed semantic case frames from a corpus [8] . There are three main differences to our approach: they use an annotated corpus, depend deeply on a thesaurus, and did not resolve verb sense ambiguity.", "cite_spans": [ { "start": 295, "end": 298, "text": "[8]", "ref_id": "BIBREF7" } ], "ref_spans": [ { "start": 135, "end": 143, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "2.", "sec_num": null }, { "text": "This section explains how to collect examples shown in Figure 1 . In order to improve the quality of collected examples, reliable modifier-head relations are extracted from the parsed corpus.", "cite_spans": [], "ref_spans": [ { "start": 55, "end": 63, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "COLLECTING EXAMPLES", "sec_num": "3." }, { "text": "When examples are collected, case markers, case examples, and case components must satisfy the following conditions.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Conditions of case components", "sec_num": "3.1" }, { "text": "Case components which have the following case markers (CMs) are collected: ga (nominative), wo (accusative), ni (dative), to (with, that), de (optional), kara (from), yori (from), he (to), and made (to). We also handle compound case markers such as ni-tsuite 'in terms of', wo-megutte 'concerning', and others.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Conditions of case markers", "sec_num": null }, { "text": "In addition to these cases, we introduce time case marker. Case components which belong to the class