翻譯20 點I. INTRODUCTION

I. INTRODUCTIONTHE goal of an information retrieval system is to evaluatethe degrees of relevance of the collected documents withrespect to a user’s queries and retrieve the documents with ahigh degree of satisfaction to the user. In order to get good retrievalperformance, the user’s query must be able to... show more I. INTRODUCTIONTHE goal of an information retrieval system is to evaluatethe degrees of relevance of the collected documents withrespect to a user’s queries and retrieve the documents with ahigh degree of satisfaction to the user. In order to get good retrievalperformance, the user’s query must be able to appropriatelydescribe the user’s requests. Currently, most of the commercialinformation retrieval systems are based on the Booleanlogic model. They assume that a user’s queries can preciselybe characterized by the index terms. However, this assumptionis inappropriate due to the fact that the user’s queries may containfuzziness [22]. The reason for the fuzziness contained in theuser’s queries is that the user may not know much about the subjecthe/she is searching or may not be familiar with the informationretrieval system. Therefore, the query specified by the usermay not describe the information request properly. Since fuzzyset theory [29] can be used to describe imprecise or fuzzy information,many researchers have applied the fuzzy set theory toinformation retrieval systems [2], [3], [8], [9], [13], [18]–[22].In [2], Bordogna et al. presented a relevance feedback modelbased on associative neural networks to provide an associationmechanism in information retrieval systems. The purpose of theassociation mechanism in information retrieval systems is tobuild the association relationships between index terms and tomodify the user’s queries by adding or replacing index termsassociated with the queries. Generally speaking, the modifieduser’s queries should find more relevant documents than thatof the original user’s queries and thus improve the retrievalperformance. Therefore, the study of the association mechanismis very important in the field of information retrieval. In[3], Chen et al. presented a fuzzy-based concept informationsystem that integrates human categorization and numerical clustering.In [8], Chen et al. presented a method for document retrievalusing knowledge-based fuzzy information retrieval techniques.In [9], Chen et al. presented fuzzy information retrievaltechniques based on multi-relationship fuzzy concept networks.In [13], Horng et al. presented a fuzzy information retrievalmethod based on document terms reweighting techniques,In [20], Kraft et al. explored several ways of using fuzzyclustering techniques in information retrieval systems, wherethe most important one is to capture the relationships amongindex terms. They use fuzzy logic rules to represent the associationrelationships between index terms and to form the basisof the association mechanism. After a user submits his/herqueries, the fuzzy logic rules are then applied under a fuzzylogic system to modify the user’s original queries. Experimentalresults show that the modified user’s queries can get abetter retrieval performance than the original queries. In [20],Kraft et al. utilized the complete link clustering method andthe fuzzy c-means clustering method to partition documentsfor information retrieval.In this paper, we extend the work of Kraft et al. [20] to presenta new method to modify a user’s queries for fuzzy informationretrieval. First, we present a fuzzy agglomerative hierarchicalclustering algorithm for clustering documents and to getthe document cluster center of each document cluster. Then, wepresent a method to construct fuzzy logic rules based on thedocument clusters and their document cluster centers. Finally,we apply the constructed fuzzy logic rules to modify the user’squery for query expansion and to guide the information retrievalsystem to retrieve documents relevant to a user’s request. Thefuzzy logic rules can represent three kinds of fuzzy relationships(i.e., fuzzy positive association relationship, fuzzy specializationrelationship, and fuzzy generalization relationship)between index terms. The proposed document retrieval methodis more flexible and more intelligent than the existing methodsdue to the fact that it can expand users’ queries for fuzzy informationretrieval in a more effective manner.The rest of this paper is organized as follows. In Section II, webriefly review some clustering methods. In Section III, we proposea new fuzzy agglomerative hierarchical clustering method.In Section IV, we compare the clustering performance of theproposed fuzzy agglomerative hierarchical clustering methodwith that of the complete link clustering method. In Section V,we propose a new method for fuzzy logic rules discovery. InSection VI, we propose a new method for query modificationfor fuzzy information retrieval based on the constructed fuzzylogic rules. The conclusions are discussed in Section VII.
Update: 英翻中,
勳雞肉 翻的"情報"應該翻成"資訊"@@
4 answers 4