{
    "paper_id": "C96-1037",
    "header": {
        "generated_with": "S2ORC 1.0.0",
        "date_generated": "2023-01-19T12:51:41.039540Z"
    },
    "title": "Aligning More Words with High Precision for Small Bilingual Corpora",
    "authors": [
        {
            "first": "Sur-Jin",
            "middle": [],
            "last": "Ker",
            "suffix": "",
            "affiliation": {
                "laboratory": "",
                "institution": "National Tsing Hua University Hsinchu",
                "location": {
                    "postCode": "30043",
                    "country": "Taiwan, ROC"
                }
            },
            "email": ""
        },
        {
            "first": "Jason",
            "middle": [
                "J S"
            ],
            "last": "Chang",
            "suffix": "",
            "affiliation": {
                "laboratory": "",
                "institution": "National Tsing tlua University Hsinchu",
                "location": {
                    "postCode": "30043",
                    "country": "Taiwan, ROC"
                }
            },
            "email": "jschang@cs.nthu.edu.tw"
        }
    ],
    "year": "",
    "venue": null,
    "identifiers": {},
    "abstract": "In this paper, we propose an algorithm for aligning words with their translation in a bilingual corpus. Conventional algorithms are based on word-byword models which require bilingual data with hundreds of thousand sentences for training. By using a word-based approach, less frequent words or words with diverse translations generally do not have statistically significant evidence for confident alignment. Consequently, incomplete or incorrect alignments occur. Our algorithm attempts to handle the problem using classbased rules which are automatic acquired from bilingual materials such as a bilingual corpus or machine readable dictionary. The procedures for acquiring these rules is also described. We found that the algorithm can align over 80% of word pairs while maintaining a comparably high precision rate, even when a small corpus was used in .training. The algorithm also poses the advantage of producing a tagged corpus for word sense disambiguation.",
    "pdf_parse": {
        "paper_id": "C96-1037",
        "_pdf_hash": "",
        "abstract": [
            {
                "text": "In this paper, we propose an algorithm for aligning words with their translation in a bilingual corpus. Conventional algorithms are based on word-byword models which require bilingual data with hundreds of thousand sentences for training. By using a word-based approach, less frequent words or words with diverse translations generally do not have statistically significant evidence for confident alignment. Consequently, incomplete or incorrect alignments occur. Our algorithm attempts to handle the problem using classbased rules which are automatic acquired from bilingual materials such as a bilingual corpus or machine readable dictionary. The procedures for acquiring these rules is also described. We found that the algorithm can align over 80% of word pairs while maintaining a comparably high precision rate, even when a small corpus was used in .training. The algorithm also poses the advantage of producing a tagged corpus for word sense disambiguation.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Abstract",
                "sec_num": null
            }
        ],
        "body_text": [
            {
                "text": "initiated much of the recent interest in bilingual corpora. They advocated applying a statistical approach to machine translation (SMT). The SMT approach can be understood as a word by word model consisting of two submodels: a language model for generating a source text segment ST and a translation model for translating ST to a target text segment TT. They recommended using an aligned bilingual corpus to estimate the parameters of translation probability, Pr(ST [TT) in the translation model. The resolution of alignment can vat3, from low to high: section, paragraph, sentence, phrase, and word (Gale and Church 1993; Matsumoto et al. 1993) .",
                "cite_spans": [
                    {
                        "start": 600,
                        "end": 622,
                        "text": "(Gale and Church 1993;",
                        "ref_id": null
                    },
                    {
                        "start": 623,
                        "end": 645,
                        "text": "Matsumoto et al. 1993)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "In addition to machine translation, many applications tbr aligped corpora have been proposed, including bilingual lexicography (Gale and Church 199l, Smadja 1992, Dallie, Gaussier and Lange 1994) , and word-sense disambiguation (Gale, Church and Yarowsky 1992, Chen and Chang 1994) .",
                "cite_spans": [
                    {
                        "start": 127,
                        "end": 136,
                        "text": "(Gale and",
                        "ref_id": null
                    },
                    {
                        "start": 137,
                        "end": 183,
                        "text": "Church 199l, Smadja 1992, Dallie, Gaussier and",
                        "ref_id": null
                    },
                    {
                        "start": 184,
                        "end": 195,
                        "text": "Lange 1994)",
                        "ref_id": null
                    },
                    {
                        "start": 228,
                        "end": 245,
                        "text": "(Gale, Church and",
                        "ref_id": null
                    },
                    {
                        "start": 246,
                        "end": 269,
                        "text": "Yarowsky 1992, Chen and",
                        "ref_id": null
                    },
                    {
                        "start": 270,
                        "end": 281,
                        "text": "Chang 1994)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "In the context of statistical machine translation, Brown et al. (1993) presented a series of five models for Pr(ST [TT) . The first two models have been used in research on word alignment. Model 1 assumes that Pr(ST[TT) depends only on lexical translation probability t(s I t), i.e., the probability of the i-th word in ST producing the j-th word t in TT as its translation. The pair of words (s, t) is called a connection. Model 2 enhances Model 1 by considering the dependence of Pr(ST ITT) on the distortion probability, d(i l J, 1, m) where I and m are the numbers of words in ST and TT, respectively.",
                "cite_spans": [
                    {
                        "start": 51,
                        "end": 70,
                        "text": "Brown et al. (1993)",
                        "ref_id": null
                    },
                    {
                        "start": 115,
                        "end": 119,
                        "text": "[TT)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "Using an EM algorithm for Model 2, Brown et al. (1990) reported the model produced seventeen acceptable translations for twenty-six testing sentences. However, the degree of success in word alignment was not reported.",
                "cite_spans": [
                    {
                        "start": 35,
                        "end": 54,
                        "text": "Brown et al. (1990)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "Dagan, Church and Gale (1992) proposed directly aligning words without the preprocessing phase of sentence alignment. Under this proposal, a rough chm'acter-by-character alignment is first performed. From this rough character alignment, words are aligned using an EM algorithm for Model 2 in a fashion quite similar to the method presented by Brown. Instead of d(i [ j, 1, m) , a smaller set of offset probabilities, o(i -i') were used where the i-th word of ST was connected to the j-th word of TT in the rough alignment. This algorithm was evaluated on a noisy English-French technical document. The authors claimed that 60.5% of 65,000 words in the document were correctly aligned. For 84% of the words, the offset from correct alignment was at most 3.",
                "cite_spans": [
                    {
                        "start": 7,
                        "end": 29,
                        "text": "Church and Gale (1992)",
                        "ref_id": null
                    },
                    {
                        "start": 350,
                        "end": 369,
                        "text": "Instead of d(i [ j,",
                        "ref_id": null
                    },
                    {
                        "start": 370,
                        "end": 372,
                        "text": "1,",
                        "ref_id": null
                    },
                    {
                        "start": 373,
                        "end": 375,
                        "text": "m)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "Motivated by the need to reduce on the memory requirement and to insure robustness in estimation of probability, Gale and Church (1991) proposed an alternative algorithm in which probabilities are not estimated and stored for all word pairs. Instead, only strongly associated word pairs are Ibund and stored. This is achieved by applying dO 2 test, a x~-like statistic. The extracted word pairs are used to match words in ST and TT. The algorithm works from left to right in ST, using a dynamic programming procedure to maximize Pr(ST ITT). The probability t(s ] t) is approximated as a function of thn-in, the number of matches (s', t) for all s' ~ ST, while distortion d(i I J, l, m) is approximated as a probability function, Pr(matchlj'-j) of slope, j'j, where (i', j') is the positions of the nearest connection to the left of s. The authors claim that when a relevant threshold is set, the algorithm can recommend connections for 61% for the words in 800 sentence pairs. Approximately 95% of the suggested connections are correct.",
                "cite_spans": [
                    {
                        "start": 113,
                        "end": 135,
                        "text": "Gale and Church (1991)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "in this paper, we propose a word-alignment algorithm based on classes derived from sense-related categories in existing thesauri. We refer to this algorithm as SenseAlign. The proposed algorithm relies on an automatic procedure to acquire classbased rules for alignment. It does not employ wordby-word translation probabilities; nor does it use a lengthy iterative EM algorithm for converging to such probabilities. Results obtained fiom the algorithms demonstrate that classification based on existing thesauri is very effective in broadening coverage while maintaining high precision. When trained with a corpus only one-tenth the size of the corpus used in Gale and Church (1991) , the algorithm aligns over 80% of word pairs with comparable precision (93%). Besides, since the rules are based on sense distinction, word sense ambiguity can be resolved in favor of the corresponding senses of rules applied in the alignment process.",
                "cite_spans": [
                    {
                        "start": 660,
                        "end": 682,
                        "text": "Gale and Church (1991)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "The rest of this paper is organized as tbllows. In the next section, we describe SenseAlign and discuss its main components. Examples of its output are provided in Section 3. All examples and their translations are taken from the l~ongman English-Chinese Dictionary of Contemporary English (Procter 1988, I,ecDOCE, hencetbrth) . Section 4 summarizes the results of inside and outside tests. In Section 5, we compare SenseAlign to several other approaches that have been proposed in literature involving computational linguistics.",
                "cite_spans": [
                    {
                        "start": 290,
                        "end": 326,
                        "text": "(Procter 1988, I,ecDOCE, hencetbrth)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "Finally, Section 6 summarized the paper.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "2. The Word Alignment Algorithm 2.1 Preliminary details. SenseAlign is a class-based word alignment system that utilizes both existing and acquired lexical knowledge. The system contains the following components and distinctive t~atures. A A greedy algorithm for aligning words. The algorithm is a greedy decision procedure for selecting preferred connections. The evaluation is based on composite scores of various factors: applicability, specificity, fan-out relative distortion probabilities, and evidence from bilingual dictionaries. B. Lexieal preprocessing. Morphological analysis, part-of-speech tagging, ktioms identification are performed for the two languages involved. In addition, certain morpho-syntactic analyses are performed to handle structures that are specific only to one of the two languages involved. By doing so, the sentences are brought closer to each other in the number of words. C. Two thesauri for classifying words. (McArthur 1992; Mei et al. 1993 ) Classification allows a word to align with a target word using the collective translation tendency of words in the same class. Class-base roles obviously have much less parameters, are easier to acquire and can be applied more broadly. 1). Two different ways of learning class-based rules. The class-based can be acquired either from bilingual materials such as example sentences and their translations or definition sentences tbr senses in a machine readable dictionary. E. Similarity between connection target and dictionary translations. In 40% of the correct connections, the target of the connection and dictionary translation have at least one Chinese character in common. To exploit this thesaury t effect in translation, we include similarity between target and dictionary translation as one of the factors. F. Relative distortion. Translation process tends to preserve contiguous syntactical structures. The target position in a connection high depends that of adjacent connections. Therelbre, parameters in an model of distortion based on absolute position are highly redundant. Replacing probabilities of the fbrm d(iLj, 1, m) with relative distortion is a feasible alternative. By relative distortion, rd for the connection (s,t), we mean (j-j')-(i-i') where i'th word, s' in the same syntactical structure of s, is connected to the j'th word, t' in TT, 2.2. Acquisition of alignment rules. Class-based alignment rules can be acquired from a bilingual corpus. Table i presents the ten rules with the highest applicability acquired from the example sentences and their translations in LecDOCE. Alternatively, we can acquire rules from the bilingual definition text for senses in a bilingual dictionary. The definition sentence are disambiguated using a sense division based on thesauri for the two language involved. Each sense is assigned codes fi'om the two thesauri according to its definition in both languages. See Table 2 lbr examples of sense definition and acquired rules.",
                "cite_spans": [
                    {
                        "start": 946,
                        "end": 961,
                        "text": "(McArthur 1992;",
                        "ref_id": null
                    },
                    {
                        "start": 962,
                        "end": 977,
                        "text": "Mei et al. 1993",
                        "ref_id": null
                    }
                ],
                "ref_spans": [
                    {
                        "start": 2452,
                        "end": 2459,
                        "text": "Table i",
                        "ref_id": null
                    },
                    {
                        "start": 2911,
                        "end": 2918,
                        "text": "Table 2",
                        "ref_id": null
                    }
                ],
                "eq_spans": [],
                "section": "Introduction",
                "sec_num": "1."
            },
            {
                "text": "Evaluation of connection candidates. Connection candidates can be evaluated using various factors of confidence. The probabilities of having a correct connection as fimctions of these fhctors are estimated empirically to reflect their relative contribution to the total confidence of a connection 1 From one aspect those words sharing common characters can be considered as synonyms tha would appear in a thesaurus. Fujii and Croft (1993) pointed out that this thesaury effect of Kanji in Japanese helps broaden tile query lhvorably for character-based information retrieval of Japanese documents. candidate, fable 3 lists the empirical probabilities of various factors. 2.4. Alignmen! algorithm.",
                "cite_spans": [
                    {
                        "start": 416,
                        "end": 438,
                        "text": "Fujii and Croft (1993)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "2.3",
                "sec_num": null
            },
            {
                "text": "Our algorithm fbr word aligmnent is a decision procedure tbr selecting the preferred connection fiom a list of candidates. The initial list of selected connection contains two dummy connections. This establishes the initial anchor points tbr calculating relative distortion. The highest scored candidate is selected and added to the list of solution. The newly added connection serves as an additional anchor for a more accurate estimation of relative distortion. The connection candidates that are inconsistent with the selected connection are removed from the list. Subsequently, the rest of the candidates are re-evaluated again. Figure 1 presents the SenseAlign algorithm.",
                "cite_spans": [],
                "ref_spans": [
                    {
                        "start": 633,
                        "end": 641,
                        "text": "Figure 1",
                        "ref_id": null
                    }
                ],
                "eq_spans": [],
                "section": "2.3",
                "sec_num": null
            },
            {
                "text": "To illustrate how SenseAlign works, consider the pair of sentences (1 e, 1 c).",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Example of running SenseAlign.",
                "sec_num": "3."
            },
            {
                "text": "( I e) I caught a fish yesterday. (lc) Zhuotian wuo budao yitiao yu. yesterday I catch one fish. Table 4 shows the connections that are considered in each iteration of the SenseAlign algorithm. Various factors used to evaluate connections are also given. Table 5 lists the connection in the final solution of alignment.",
                "cite_spans": [],
                "ref_spans": [
                    {
                        "start": 97,
                        "end": 104,
                        "text": "Table 4",
                        "ref_id": null
                    },
                    {
                        "start": 255,
                        "end": 262,
                        "text": "Table 5",
                        "ref_id": "TABREF1"
                    }
                ],
                "eq_spans": [],
                "section": "Example of running SenseAlign.",
                "sec_num": "3."
            },
            {
                "text": "In this section, we present the experimental results of an implementation of SenseAlign and related algorithms. Approximately 25,000 bilingual example sentences from LecDOCE are used here as the training data. Here, the training data were used primarily to acquire rules by a greedy learner and to determine empMcally probability thnctions of various factors. The algorithnfs pertbrmance was then tested on the two sets of reside and outside data. The inside test consists of fitty sentence pairs from LecDOCE as input. The outside test are 416 sentence pairs fiom a book on English sentence patterns containing a comprehensive fifty-five sets of typical sentence patterns, l lowever, the words in this outside test is somewhat more common, and, thereby, easier to align. \"fhis is evident from the slightly higher hit rate based on simple dictionary lookup. The first experiment is designed to demonstrate the effectiveness of an naive algorithm (DictAlign) based on a bilingual dictionary. According to our results, although DictAlign produces high precision alignment the coverage for both test sets is below 20%. However, if the thesaury eft}act is exploited, the coverage can be increased nearly three tblds to about 40%, at the expense of a decrease around 10% in precision, ",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Experiments with SenseAlign",
                "sec_num": "4."
            },
            {
                "text": "V I Di Dc098 lhn05 l-1 0 15,3 ~i-Y 0 (] 0 0 0 0 0 01 0 0[ 0 0 i 0 ()i o o! 3 I PP :J:~ Nh Gh280 Na02 I-1 I 0 () ((~ 3 I PP 4-~ Nh Gh280 Na05 1-I I 0 0 3 a AT -{~ Ne Nd098 Qa04 1-1 0.5 0 0 0 3 caught VB 41115i~j V+Di Dc098 lhn05 1-I 0 0 0 0: 4 a AT \"ti~ Nc Nd098 Qa04 1-I 0.5 0 0 () 4 caught VB l][]j!I] V ~.[)i 1)e098 Hm05 l-1 0 0 0 5 caught VB ~II] 5_!r] V+Di De098 Hm05 l -1 0 0 0 0",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Experiments with SenseAlign",
                "sec_num": "4."
            },
            {
                "text": "In our second experiment, we use SenseAlign described above for word aligmnent except that no bilingual diclionary is used. In our thiM expet+iment, we use the full SenseAlign to align the testing data. Table 6 indicates that acquired lexical infornmtion augmented and existing lexical information such as a bilingual dictionary can supplement each other to produce optimum aligmnent results. The generality of the approach is evident fi-om the thct that the coverage and precision for the ovtside test are comparable with those of the inside test.",
                "cite_spans": [],
                "ref_spans": [
                    {
                        "start": 203,
                        "end": 210,
                        "text": "Table 6",
                        "ref_id": null
                    }
                ],
                "eq_spans": [],
                "section": "Experiments with SenseAlign",
                "sec_num": "4."
            },
            {
                "text": "We believe the proposed algorithm addresses tile problem of knowledge engineering bottleneck by using both corpora and machine readable lexical resources such as dictionaries and thesauri. The corpora provide us with training and testing materials, so that empirical knowledge can be derived and evaluated objectively. The thesauri provide classification that can be utilized to generalize the empirical knowledge gleaned fi-om corpora SenseAlign achieves a degree of generality since a word pair can be accurately aligned, even when they occur rarely or only once ill the corpus. This kind of generality is unattainable by statistically trained wordbased lnodels. Class-based models obviously offer advantages of smaller storage requirement and hi vher system efficiency. Such advantages do have their costs, tot' class-based models may be over-generalized and miss word-specific rules. However, work on classbased systems have indicated that the advantages oulweigh the disadvantages. 5.2 Mutual information, and frequency. Gale and Church (1990) shows a near-miss example where (]2 a Z2-1ike statistic works better than mutual infimnation for selecting strongly associated woM pairs to use in word alignment. In their study, they contend that 2 like statistic works better because it uses co-nonoccurrence and the number of sentences where one word occurs while the other does not which are often larger, more stable, and more indicative than cooccurrence used in mutual information.",
                "cite_spans": [
                    {
                        "start": 1026,
                        "end": 1048,
                        "text": "Gale and Church (1990)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Machine-readable lexieal resources vs. corpora",
                "sec_num": "5.1"
            },
            {
                "text": "The above-cited work's discussions of the Z2-1ike statistic and the fan-in factor provide a valuable reference for this work. In our attempt to improve on low coverage of word-based approaches, we use simple filtering according to fan-out in the acquisition of class-based rules, in order to maximize both coverage and precision. The rules that provide the most instances of plausible connection is selected. This contrasts with approaches based on wordspecific statistic where strongly associated word pairs selected may not have a strong presence in the data. This generally corresponds to the results from a recent work on a variety of tasks such as terminology extraction and structural disambiguation. Dallie, Gaussier and Lange (1994) demonstrated that simple criteria related to frequency coupled with a linguistic filter works better than mutual information tbr terminology extraction. Recent work involving structural disambiguation (Brill and Resnik 1994) also indicated that statistics related to frequency outperform mutual intbrmation and q~2 statistic.",
                "cite_spans": [
                    {
                        "start": 707,
                        "end": 740,
                        "text": "Dallie, Gaussier and Lange (1994)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Machine-readable lexieal resources vs. corpora",
                "sec_num": "5.1"
            },
            {
                "text": "This paper has presented an algorithm capable of identit~,ing words and their translation in a bilingual corpus. It is effective for specific linguistic reasons. The significant majority of words in bilingual sentences have diverging translation; those translations are not often tbund in a bilingual dictionaly. However, those deviation are largely limited within the classes defined by thesauri. Therefore, by using a class-based approach, the problem's complexity can reduced in the sense that less number of candidates need to be considered with a greater likelihood of finding the correct translation.",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Concluding remarks",
                "sec_num": "6."
            },
            {
                "text": "In general, a slight amotmt of precision can apparently be expended to gain a substantial increase in applicability. Our results suggest that mixed strategies can yield a broad coverage and high precision word alignment and sense tagging system which can produce richer information fbr MT and NLP tasks such as word sense disambiguation. The word sense information can provide a certain degree of generality which is lacking in most statistical procedures. The algorithm's performance discussed here can definitely be improved by enhancing the various components of the algorithm, e.g., morphological analyses, bilingual dictionary, monolingual thesauri, and rule acquisition. However, this work has presented a workable core for processing bilingual corpus. The proposed algorithm can produce effective word-alignment results with 1. Read a pair of English-Chinese sentences. 2. Two dummies are replace to the left of the first and to the right of the last word of the source sentence. Similar two dummies are added to the target sentence. The left dummy in the source and target sentences align with each other. Similarly, the right dummies align with each other. ]'his establishes anchor points for calculating the relative distortion score. 3. Perfbrm the part-of-speech tagging and analysis tbr sentences in both languages. 4. Lookup the words in LEXICON and C1LIN to determine the classes consistent with the partof-speech analyses. 5. Follow the procedure in Section 2.3 to calculate a composite probability tbr each connection candidate according to fan-out, applicability, specificity of alignment rules, relative distortion, and dictionary evidence. 6. The highest scored candidate is selected and added to the list of alignment. 7. The connection candidates that are inconsistent with the selected connection are also removed from the candidate list. 8. The rest of the candidates are evaluated again according to the new list of connections. 9. The procedure iterates until all words in the source sentence are alibi. Figure 1 . Alignment Algorithm of SenseAlign sense tagging which can provide a basis for such N I,P tasks as word sense disambiguation (Chen and Chang 1994) and PP attachment (Chen and Chang 199\"5) .",
                "cite_spans": [
                    {
                        "start": 2205,
                        "end": 2227,
                        "text": "(Chen and Chang 199\"5)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [
                    {
                        "start": 2030,
                        "end": 2038,
                        "text": "Figure 1",
                        "ref_id": null
                    }
                ],
                "eq_spans": [],
                "section": "Concluding remarks",
                "sec_num": "6."
            },
            {
                "text": "While this paper has specifically addressed only English-Chinese corpora, the linguistic issues that motivated the algorithm are quite general and are to a great degree language independent. If such a case is true, the algorithm presented here should be adaptable to other language pairs. The prospects tbr Japanese, in particular, seem highly promising There arc some work on alignment of l\u00a3nglish-Japanese texts using both dictionaries and statistics (Utsuro, lkeda, Yamane, Matsumoto and Nagao 1994) .",
                "cite_spans": [
                    {
                        "start": 453,
                        "end": 502,
                        "text": "(Utsuro, lkeda, Yamane, Matsumoto and Nagao 1994)",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Concluding remarks",
                "sec_num": "6."
            }
        ],
        "back_matter": [
            {
                "text": "The authors would like to thank the National Science Concil of the Republic of China for financial support of this manuscript under Contract No. NSC 84-102-",
                "cite_spans": [],
                "ref_spans": [],
                "eq_spans": [],
                "section": "Acknowledgments",
                "sec_num": null
            },
            {
                "text": "1211. Zebra Corporation and l,ongnmn Group arc appreciated tbr the machine readable dictionary. Special thanks are due to Mathis H. C Chen lbr work of preprocessing the Mill). Thanks are also due to Keh-Yih Su tbr many helpful comments on an early drall of this paper. References 1. Brill, Eric and P. Resnik, (1994) . A Rule bascd Approach to l'repositional Phrase Attachment, In l~roceedings oJ\" Ihe 15lh hJlernaliotml (\"ot!/L, rence on ( 7omlmlalional Linguistics, [ 198-1205 Trattsktlion, 101-112, Montreal, CanadaKay, Martin and Martin Iloscheisen, (1993) . Text-Translation Aligmnent, Computational Linguistics, Vol. 19, No. 1, page 121-142. 12.1,ongman, (1993) . lxmgman English-Chinese Dictionary of Contemporary English, Published by l,ongnmn Group (Far l,;ast) I,td., I long Kong. 13.Matsumoto, Y. et al. (1993) Nagao, (1994) . Bilingual text matching using bilingual dictionary anti statistics, In l~roceedin\u00a3<~ ' (?]\" the 151h Inlernational ( ?oql'ereuce on ('~ompulaliona[ Linguistics, page 1076 ( ?oql'ereuce on ('~ompulaliona[ Linguistics, page -1083 , Kyoto, Japan.9.",
                "cite_spans": [
                    {
                        "start": 290,
                        "end": 316,
                        "text": "Eric and P. Resnik, (1994)",
                        "ref_id": null
                    },
                    {
                        "start": 421,
                        "end": 478,
                        "text": "(\"ot!/L, rence on ( 7omlmlalional Linguistics, [ 198-1205",
                        "ref_id": null
                    },
                    {
                        "start": 479,
                        "end": 560,
                        "text": "Trattsktlion, 101-112, Montreal, CanadaKay, Martin and Martin Iloscheisen, (1993)",
                        "ref_id": null
                    },
                    {
                        "start": 563,
                        "end": 667,
                        "text": "Text-Translation Aligmnent, Computational Linguistics, Vol. 19, No. 1, page 121-142. 12.1,ongman, (1993)",
                        "ref_id": null
                    },
                    {
                        "start": 785,
                        "end": 821,
                        "text": "Kong. 13.Matsumoto, Y. et al. (1993)",
                        "ref_id": null
                    },
                    {
                        "start": 822,
                        "end": 835,
                        "text": "Nagao, (1994)",
                        "ref_id": null
                    },
                    {
                        "start": 953,
                        "end": 1008,
                        "text": "( ?oql'ereuce on ('~ompulaliona[ Linguistics, page 1076",
                        "ref_id": null
                    },
                    {
                        "start": 1009,
                        "end": 1065,
                        "text": "( ?oql'ereuce on ('~ompulaliona[ Linguistics, page -1083",
                        "ref_id": null
                    }
                ],
                "ref_spans": [],
                "eq_spans": [],
                "section": "annex",
                "sec_num": null
            }
        ],
        "bib_entries": {},
        "ref_entries": {
            "TABREF0": {
                "content": "<table><tr><td/><td colspan=\"2\">, Ten rules with the</td><td colspan=\"3\">highest applicability</td></tr><tr><td>#~</td><td>Rule</td><td/><td colspan=\"3\">Gloss lbr classes</td></tr><tr><td>I</td><td colspan=\"2\">642 Ma001, Hj63</td><td colspan=\"3\">moving / come, and go</td></tr><tr><td>2</td><td colspan=\"2\">459 ,lh210. Dil9</td><td colspan=\"3\">jobs, trade / work</td></tr><tr><td>3</td><td colspan=\"2\">440 Md108, Be21</td><td>trams/car</td><td/></tr><tr><td>4</td><td colspan=\"2\">4t8 L8202 , Eb28</td><td colspan=\"3\">new/ne~, flesh</td></tr><tr><td>5</td><td colspan=\"2\">367 DaO03, Bn01</td><td colspan=\"3\">huildmg, house/building</td></tr><tr><td>6</td><td colspan=\"2\">362 Gc060, Hil6</td><td colspan=\"3\">spcaking/ introduce</td></tr><tr><td>7</td><td colspan=\"2\">349 Fc050, Ed03</td><td colspan=\"3\">qualities / good. bad</td></tr><tr><td>8</td><td colspan=\"2\">310 Lh226, TII 8</td><td>nicotine</td><td/><td>/ time</td></tr><tr><td>9</td><td colspan=\"2\">303 Ca002, Ab04</td><td colspan=\"3\">man and ~oman / babx</td></tr><tr><td>1o</td><td colspan=\"2\">302 'Fb020. Gb09</td><td colspan=\"3\">likin% loving / like, love</td></tr><tr><td colspan=\"6\">Table 2. Rules acquired from bilingual definitions for 12</td></tr><tr><td colspan=\"3\">senses of\"bank'\" m LDOCIE.</td><td/><td/></tr><tr><td/><td colspan=\"2\">SgllbC &amp; 1)cfinition</td><td/><td/><td>Rules</td></tr><tr><td colspan=\"4\">I I.n. 11 land along the side of a river, lake</td><td/><td>1,d099, 13e03</td></tr><tr><td>etc.</td><td>i'/: ; ~i~</td><td/><td/><td/></tr><tr><td colspan=\"4\">II .n.2] earth which is heaped up m a field or</td><td/><td>I x1{}99. Bn 12</td></tr><tr><td colspan=\"4\">garden, often making a border or division.</td><td/></tr><tr><td colspan=\"2\">II IJ:~</td><td/><td/><td/></tr><tr><td colspan=\"4\">l.n.3] a mass ofsnmv, clouds, mud, etc. -</td><td/><td>Ib, Bb(i3</td></tr><tr><td colspan=\"2\">~I[i : -l' Jfl</td><td/><td/><td/></tr><tr><td colspan=\"5\">I.n.4[ a slope made at bends in a road ori</td><td>Ix10!}9. I ~c(},}</td></tr><tr><td colspan=\"4\">race-track, so that they are safer for cars tc</td><td/></tr><tr><td colspan=\"2\">go round. ~]J~'</td><td/><td/><td/></tr><tr><td colspan=\"3\">l.n.51 = SANDBANK. , f,~ll'l'\";</td><td/><td/><td>I,d(}99, I~c02</td></tr><tr><td colspan=\"4\">12.v. II (of a car or aircraft) to move ~ith</td><td/><td>N.j295, I:d{}2</td></tr><tr><td colspan=\"5\">one side higher than the other, esp. Mama</td></tr><tr><td colspan=\"3\">making a turn {~' [$;I.qi~g,~</td><td/><td/></tr><tr><td colspan=\"4\">il3 n. I I a row, csp. of OARs in an ancient</td><td colspan=\"2\">l Ib, I )n{}8</td></tr><tr><td colspan=\"4\">boat or KEYs on a TYPEWRfFER ~Uf</td><td/></tr><tr><td colspan=\"4\">14.n. II a placc in which money is kept and</td><td colspan=\"2\">Jel04, l)m0,1</td></tr><tr><td colspan=\"4\">paid out on demand, and where related</td><td/></tr><tr><td colspan=\"3\">aetMties go on. ,JI[47 j:</td><td/><td/></tr><tr><td colspan=\"4\">14.n.2i (usu. in comb.) a place where</td><td colspan=\"2\">.cI04 Ih~17</td></tr><tr><td colspan=\"4\">something is held ready for use, esp.</td><td/></tr><tr><td colspan=\"4\">ORGANIC producls o1\" lmman origin for</td><td/></tr><tr><td colspan=\"3\">medical use. {i~{(igi!( \"</td><td/><td/></tr><tr><td colspan=\"4\">14.n.31 (a person Mlo keeps) a supply of</td><td colspan=\"2\">.le104. I )m04</td></tr><tr><td colspan=\"4\">moncy or pieces for paymcnt or use in a</td><td/></tr><tr><td colspan=\"3\">game of chance. ;~p~</td><td/><td/></tr><tr><td colspan=\"4\">15.v. 11 to put or kcep (money) in a bank (%:</td><td colspan=\"2\">Jc 106. I Ij,l{}</td></tr><tr><td colspan=\"6\">15.v.21 [csp. with] to keel} one's money .lelO6,1lj40</td></tr><tr><td colspan=\"4\">!esp. in the stated bank) (/di\u00a3</td><td/></tr><tr><td colspan=\"6\">Table 3. Factor types with empirical probability</td></tr><tr><td colspan=\"2\">Factor</td><td colspan=\"2\">condition and probability</td><td/></tr><tr><td>Fo</td><td>f= 1</td><td>/\": 2</td><td colspan=\"3\">f= 3</td><td>f&gt; 3</td></tr><tr><td>Prob</td><td>0.85</td><td>0.61</td><td colspan=\"3\">0.44</td><td>0.42</td></tr><tr><td>App</td><td>A &gt;. 1</td><td>\u2022 I&gt;A k .01</td><td colspan=\"2\">.0I&gt;A &gt; .001</td><td>10-% A</td></tr><tr><td>Prob</td><td>0.95</td><td>0.90</td><td colspan=\"3\">0.85</td><td>0.43</td></tr><tr><td>Spec</td><td>3' _&gt; 12</td><td>12&gt;3\"#11</td><td colspan=\"3\">11&gt; S_&gt;I0</td><td>'10 &gt; ,S</td></tr><tr><td>Prob</td><td>0.95</td><td>0.85</td><td colspan=\"3\">0.77</td><td>0.35</td></tr><tr><td>R.D.</td><td>rd = 0</td><td>rd :: I</td><td colspan=\"3\">rd = 2</td><td>rd&gt; 2</td></tr><tr><td>Prob</td><td>0.26</td><td>O. 1 I</td><td colspan=\"3\">0.07</td><td>0.04</td></tr><tr><td>Sire</td><td>3'ira = I</td><td>1 =Sire&gt;.66</td><td colspan=\"3\">.66~57m ~,2</td><td>bhm &lt; .2</td></tr><tr><td colspan=\"2\">eTg-,7 o.94</td><td>0.42</td><td colspan=\"3\">0.35</td><td>0.12</td></tr></table>",
                "html": null,
                "text": "",
                "num": null,
                "type_str": "table"
            },
            "TABREF1": {
                "content": "<table><tr><td>.</td><td colspan=\"3\">q'he final alignment</td><td/></tr><tr><td>English</td><td>English</td><td/><td colspan=\"2\">Chinese</td><td>Chinese</td></tr><tr><td>Word</td><td>Code</td><td/><td>Word</td><td/><td>Code</td></tr><tr><td>I</td><td>Gh280</td><td/><td>wuo</td><td/><td>Na05</td></tr><tr><td>caught</td><td>I)e098</td><td colspan=\"2\">bu-dao</td><td/><td>Hm05</td></tr><tr><td>i a</td><td>Nd098</td><td colspan=\"2\">yi-tiao</td><td/><td>Qa04</td></tr><tr><td>!fish</td><td>Ab032</td><td colspan=\"2\">iYU</td><td/><td>Bil4</td></tr><tr><td>yesterday</td><td>Lh225</td><td/><td>zuotian</td><td/><td>Tq23</td></tr><tr><td colspan=\"4\">Table 6. Experimental Results</td><td/></tr><tr><td>\"</td><td colspan=\"3\">Inside Test</td><td/></tr><tr><td colspan=\"2\">No. Matched</td><td/><td colspan=\"3\"># ftk Coverage</td><td>Pmcision</td></tr><tr><td colspan=\"2\">DictAlign with sim = 1.0</td><td>59</td><td>56</td><td/><td>15.3%</td><td>94.9%</td></tr><tr><td colspan=\"3\">I DictAlign withsim &gt; 0.67 113</td><td>100</td><td/><td>29.4%</td><td>88.5%</td></tr><tr><td colspan=\"2\">Dici/klign with sire &gt; 0.5</td><td>l 51</td><td>124</td><td/><td>39.2%</td><td>82.1%</td></tr><tr><td colspan=\"6\">ScnseAlign wilhout sire 237 213 61.7%</td><td>89.9%</td></tr><tr><td colspan=\"2\">Full ScnseAlign</td><td colspan=\"4\">314 293 81.8%</td><td>93.3%</td></tr><tr><td/><td colspan=\"3\">Outside Test</td><td/></tr><tr><td colspan=\"2\">No. Matched</td><td/><td>fl Hit</td><td colspan=\"2\">(~ovcta~c</td><td>Prccisi()n</td></tr><tr><td colspan=\"2\">DictAlign with sire : 1.0</td><td>499</td><td>486</td><td/><td>16.8%</td><td>97.4%</td></tr><tr><td colspan=\"3\">DictAlign with sin; &gt; 0.67 970</td><td>865</td><td/><td>32.7%</td><td>89.2%</td></tr><tr><td colspan=\"2\">:DictAlign with sim &gt; 0.5</td><td>1221</td><td>1046</td><td/><td>41.1%</td><td>85.7%</td></tr><tr><td colspan=\"2\">SenscAlign wifllout sire</td><td>1913</td><td>1721</td><td/><td>66.8%</td><td>90.0%</td></tr><tr><td colspan=\"2\">:Full SenseAlign</td><td>2424</td><td>2265</td><td/><td>84.7%</td><td>93.4%</td></tr></table>",
                "html": null,
                "text": "",
                "num": null,
                "type_str": "table"
            }
        }
    }
}