File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/metho/92/c92-2104_metho.xml
Size: 15,052 bytes
Last Modified: 2025-10-06 14:13:01
<?xml version="1.0" standalone="yes"?> <Paper uid="C92-2104"> <Title>Learning Mechanism in Machine Translation System &quot;PIVOT&quot;</Title> <Section position="4" start_page="0" end_page="0" type="metho"> <SectionTitle> 2. Analysis Editing Function </SectionTitle> <Paragraph position="0"> The user can interaetively ~peeify the following information related to dependency relation by using analysis editing function of P1VOT/JE.</Paragraph> <Paragraph position="1"> (I) Dependencg (syntactic dependent end syntactic head) (2) Case (3) Parallel (4) Scope ACRES DE COLING-92, NANTES, 23-28 AOtJT 1992 6 9 3 PROC. OF COTING-92, NAt.S, AUO. 23-28, 1992 (5) Sharing The dependency relation which the system analyzes is displayed on the screen as shown in Figure 1. An underline is drawn under each Japanese phrase (a word ~itb s particle). The dependency is shown by the line which connects tmo phrases. The thick line indicates the dependency corrected by the user. Case is displayed on the line of the dependency in the form of the particles which have one-to-one correspondence with one of the cases. The bo~ indicates the correct case specified by the user. The user directly corrects above-mentioned information by using a louse and carries out translation operation once again. The translation rule controls the analysis to reflect the correction by the</Paragraph> <Section position="1" start_page="0" end_page="0" type="sub_section"> <SectionTitle> 2.1 Dependency </SectionTitle> <Paragraph position="0"> The user can correct dependency. In Figure 2, syntactic head of &quot;~--~(user)&quot; is changed from &quot;~1~ )dT~-~ (analyze)&quot; to &quot;~r~31&quot;~ (specify)&quot;.</Paragraph> <Paragraph position="1"> uAer analyze necessary Infor=|tJo~ speclf~</Paragraph> </Section> <Section position="2" start_page="0" end_page="0" type="sub_section"> <SectionTitle> 2.2 Case </SectionTitle> <Paragraph position="0"> Case shows the semantic relation between two phrases which are in dependency relation. PIVOT has more than fort~ kinds of eases such as Agent and Reason. On the screen, particles are used to express cases.</Paragraph> <Paragraph position="1"> In Figure 3, the case between &quot;EWS4800&quot; and &quot;11~31&quot; ~(run)&quot; is changed froa &quot;Contents&quot; to &quot;Place&quot; .</Paragraph> <Paragraph position="3"/> </Section> <Section position="3" start_page="0" end_page="0" type="sub_section"> <SectionTitle> 2.3 Parallel </SectionTitle> <Paragraph position="0"> The user can specify the information that two phrases are in parallel relation, Because parallel relation is one of the PIVOT eases, this function enables the user to correct dependency and case at the s~e time.</Paragraph> </Section> <Section position="4" start_page="0" end_page="0" type="sub_section"> <SectionTitle> 2.4 Scope </SectionTitle> <Paragraph position="0"> The user can specify scope. Scope means the phrase sequence in which only the syntactic head has dependency relation with other phrases outside of it.</Paragraph> </Section> <Section position="5" start_page="0" end_page="0" type="sub_section"> <SectionTitle> 2.5 Sharing </SectionTitle> <Paragraph position="0"> In Figure 1, &quot;~(user)&quot; is the subject of &quot;~ (specify)&quot; and at the same time it is the subJect of &quot;l~(translate)&quot;. In such a case, we say&quot;user&quot; is shared by &quot;~ (specify)&quot; and &quot;~'C/-$ (translate)&quot;.</Paragraph> <Paragraph position="1"> Specification of sharing is done by specifying more than one syntactic heads for the dependent. So the sharing is decomposed into dependency relations.</Paragraph> <Paragraph position="2"> Useful information on dependency relation is gotten from the user's specification of scope and so on, but this paper discusses learning from correction operation for dependency and case onlY.</Paragraph> </Section> </Section> <Section position="5" start_page="0" end_page="0" type="metho"> <SectionTitle> 3. Learning Mechanism </SectionTitle> <Paragraph position="0"> Proposed learning mechanism is as follows.</Paragraph> <Section position="1" start_page="0" end_page="0" type="sub_section"> <SectionTitle> 3.1 Learning Process </SectionTitle> <Paragraph position="0"> (1) PIVOT analyzes a source sentence.</Paragraph> <Paragraph position="1"> (2) PIVOT displays the analysis result.</Paragraph> <Paragraph position="2"> (3) A user corrects mistakes in the analysis result. (4) After the user finishes asking corrections, PIVOT translates the sentence again.</Paragraph> <Paragraph position="3"> (5) PIVOT asks the user whether translation has been a success or not, (O) If the translation is s success, PIVOT stores the analysis result together with the instruction item into an association database. If the translation is a failure, PIVOT does nothing further.</Paragraph> <Paragraph position="4"> 3.2 Applying Process (I) PIVOT analyzes a source sentence, (2) If there is ambiguity at s certain stage of analysis, PIVOT retrieves data in the association database.</Paragraph> <Paragraph position="5"> (3) PIVOT compares the possible analysis structures of the given sentence with the analysis results accumulated in the association database.</Paragraph> <Paragraph position="6"> (4) PIVOT selects the analysis structure that matches with the analysis results accumulated in the association database. If no matching occurs, PIVOT selects one structure by further application of the analysis rules.</Paragraph> <Paragraph position="7"> PIVOT learns correct analysis structures related to user's instruction. The smallest unit of PIVDT's analysis structure, that is, the triplet of syntactic dependent (with particles and voice information), syntactic head (with voice information), and the ease ACRES bE COUNG-92, NAN'IT.S, 23-28 AOt3&quot;l&quot; 1992 6 9 4 PROC. OF COLING-92, NAI~'rEs. AUG. 2.3-28, 1992 betmeen them. combined with the instruction item forms the learning unit. The instruction item shoms what the correction has been made on, namely, case or dependency correction. Each learning unit is accumulated in the association database. The database nan be retrieved mith the spelling of the syntactic dependent or head as the key. The learning unit corresponds to the follol'ing structure.</Paragraph> <Paragraph position="8"> Example of the learning process and the applying process is shomn below. This is the exaaple of correcting dependency.</Paragraph> <Paragraph position="9"> \[Translation process at the first stage\]</Paragraph> <Paragraph position="11"> he he opera glasses I</Paragraph> <Paragraph position="13"> opera l~lasses INS: Instrument If there is no information in the association database, analysis structure 1 is selected by further application of the rules.</Paragraph> <Paragraph position="14"> Translated sentence: He looked at the man who is singing with opera glasses.</Paragraph> <Paragraph position="15"> \[Instruction by User and the Learning Process\] The user corrects the analysis results. Correction of dependency: The user changes the syntactic head of &quot;:t~,'~q'92~ (opera glasses)&quot; from &quot;{1~-9~;5 (sing)&quot; to &quot;~.~ (look).&quot; Translated sentence: lie looked at a singing man with opera glasses, Learning: PIVOT stores the correct analysis structure with dependency as the instruction itea in the association database.</Paragraph> </Section> <Section position="2" start_page="0" end_page="0" type="sub_section"> <SectionTitle> 3.3 Watching Methods </SectionTitle> <Paragraph position="0"> The learning mechanism decreases the number of user's instructions. The problem is to find the effective matching method in the learning mechanism.</Paragraph> <Paragraph position="1"> Ie made experiments on four types of matching methods and compared the efficiency of each method.</Paragraph> <Paragraph position="2"> The matching methods are: (1) Restricted exact matching (2) Non-restricted exact matching (3) Restricted best latching Ac'rEs DE COLING-92. NANTES, 23-28 AO~r 1992 6 9 5 PROC. OF COLING-92. NANTEs. AUG. 23-28. 1992 (4) Non-restricted best matching Restricted exact matching is a well-known method. This method is used in many fields now. There is no study about non-restricted exact watching. Restricted best watching is a comparatively new aethod. Experiment by Wiura\[4\] is the first. There is no study about non-restricted best satchinC/.</Paragraph> <Paragraph position="3"> In restricted matching, the item in applying process has to be the same with the instruction item in learning. When the items are different, PIVOT will not use learned data. For example, if the instruction item in learning is case, PIVOT will use the learned correct analysis structure only for case selection. It will not use the data for selection of dependency or translation equivalent of each word.</Paragraph> <Paragraph position="4"> In non-restricted matching, the item in applying process need not be the same with the instruction item in learning. For example, if the instruction itew in learning is case, PIVOT will use this learned data for selection of dependency and translation equivalent of each word as well.</Paragraph> <Paragraph position="5"> The difference between the actions of restricted matching and non-restricted matching is described belo*. Consider a sentence mith two possible analysis structures. null</Paragraph> <Paragraph position="7"> Assume the following analysis structure is already learned by correcting case.</Paragraph> <Paragraph position="9"> Using restricted matching, the system selects structure 1 with its usual analysis procedure. In this case, data learned by case correction cannot be used in selection of dependenc~. Using non-restricted matching, the system selects structure R, because the learned pattern matches with the part of structure 2.</Paragraph> <Paragraph position="10"> Exact matching makes matching only once. while best matching makes matching several times. Best matching is also called associative reasoninC/.</Paragraph> <Paragraph position="11"> The difference of actions between the two methods is illustrated below.</Paragraph> <Paragraph position="13"> CASEI bet (CI,KR,Wl) stand for the learned / structure as shown on the left.</Paragraph> <Paragraph position="14"> wordl(dependent) Suppose that the following data is accumulated in the association database through dependency instructions. null There are two possible syntactic heads, W7 and 15, for W2.</Paragraph> <Paragraph position="15"> \[Action\] First, the association database is searched for patterns (x,W7,W2) and (~.|5,W2). (x:don't care)</Paragraph> </Section> <Section position="3" start_page="0" end_page="0" type="sub_section"> <SectionTitle> Database Search pattern Batching </SectionTitle> <Paragraph position="0"/> <Paragraph position="2"> In this case, there is no data that exactly matches</Paragraph> </Section> </Section> <Section position="6" start_page="0" end_page="0" type="metho"> <SectionTitle> A(.TES DE COLING-92, NANrl~, 23-28 AOt~'r 1992 6 9 6 Pgoc. o1: COLING-92, NANTES, AUG. 23-28, 1992 </SectionTitle> <Paragraph position="0"> with search patterns. However, there is data (C3,W3,\[2) that matches mith syntactic dependent. The system retrieves more information in the database so as to decide mhich of W5 and W7 is more similar to W3.</Paragraph> <Paragraph position="1"> Searching database for patterns (=,x,W3) and (x,W3,*), the following data is obtained.</Paragraph> <Paragraph position="3"> Searching database for patterns (*,*,WT) and (*,WT,*), the following data is obtained.</Paragraph> <Paragraph position="4"> (C4,W3,W7) Let this set of data be called (C3,W5,W7) &quot;database(W7).&quot; Searching database for patterns (~,~,W5) and (=.W5,x), the following data is obtained.</Paragraph> <Paragraph position="5"> (C3,W5.WT) Let this set of data be called (C1,WB,W1) &quot;database(15).&quot; On the assumption that W3 is tbe same as W7, the system performs exact matching between database(W3) and database(W7). In the following, \[W3\] is regarded as WT. On the assumption that W3 is the same as W5, the system performs exact matching between database(W3) and database(WB). In the following, \[W3\] is regarded as WS. Because the number of matches between database(W3) and databaso(WB) is larger than that between datebase(W3) and database(W7), W5 is considered to be more similar to W3 than W7. IS is selected as the head. Let PDBi(PCi,PHi,PHi.PTi) (l<=i<=n) be a possible analysis structure, where PCi: Case. PHi: Head, PDi:Bependent, PTi:Item. PDB is called &quot;possible analysis structures database&quot;. Let ADBk(ACk,AHk,ADk.ATk) (l<=k<=m) be an association database entry, xhere ACk: Case, AHk: Head, hOk:Dependent, ITk:Item. ADB is called &quot;association database&quot;. Matching algorithm for dependency selection is shown belom. All PDi's in PDH are supposed to be the same and lost of PCi's in PDB are supposed to be &quot;don't care&quot; for ease of understanding.</Paragraph> <Section position="1" start_page="0" end_page="0" type="sub_section"> <SectionTitle> First Step: </SectionTitle> <Paragraph position="0"> Extract all hDBk's such that PDi==AHk(l<=i<=n, l<=k< =m) from ADB and create SADBj(SCj,SHj,SDj,STj) (l<=j<= p), mhere SCJ: Case, SHj: Head, SDJ:Dependent, STj:Item. SADB is a subset of ADH.</Paragraph> <Paragraph position="1"> If nothing is in SADB, stop search and return fail.</Paragraph> </Section> <Section position="2" start_page="0" end_page="0" type="sub_section"> <SectionTitle> Second Step: </SectionTitle> <Paragraph position="0"> (l)Rostricted exact matching Let WORK be an empty database.</Paragraph> <Paragraph position="1"> for i=l to n for j=l to p if (SCj::PCi & SHj==Ptli & STj=:PTi) then add PDBi to WORK; endif end end return WORK; (2)Hoe-restricted exert matching Let WORK be an empty database. for i=l to n for j=l to p if (SCj==PCi & SHj==PHi) then add PDBi to WORK; endif end end return WORK; (3)Restricted best matching Let WBRKI, WORK2 be empty databases. cnt=O; * for i=1 to n for j=l to p if (SCj==PCi & SHim=PHi & STj==PTi) thee add POBi to WflRKI; exact matching between X and Y; Let cntl be the number of matched entries between X and Y; if (cntl>O & cntl==cnt) then add PDBi to WORK2; The algorithm is the same as (3) except that non-restricted exact matching is performed between X and Y instead of restricted exact matching. In the above, if more than one entries are in WORK or WORX1, the system mill select one that is most recently stored by the user's instruction. If WORX2 has more than one entries, one entry will be selected by further application of the rules. Watching algorithm for case selection is similar to that for dependency selection.</Paragraph> </Section> </Section> class="xml-element"></Paper>