DETERMINATION PROGRAM, DETERMINATION DEVICE AND DETERMINATION METHOD
PROBLEM TO BE SOLVED: To improve extraction accuracy of a synonym or a polysemic word.SOLUTION: An information processing device 1 accepts a plurality of sentences and designation of terms included in each of the plurality of sentences, and semantically analyzes each of the accepted plurality of sen...
Gespeichert in:
Hauptverfasser: | , , |
---|---|
Format: | Patent |
Sprache: | eng ; jpn |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | PROBLEM TO BE SOLVED: To improve extraction accuracy of a synonym or a polysemic word.SOLUTION: An information processing device 1 accepts a plurality of sentences and designation of terms included in each of the plurality of sentences, and semantically analyzes each of the accepted plurality of sentences to generate, for each accepted designated term, semantic structure 22 between the term and each of a plurality of other terms included in the sentence having the term. The information processing device 1 extracts, for each accepted designated term, specific semantic structure 22 from the generated semantic structure 22, and generates a feature vector 23 originated from the extracted semantic structure 22. The information processing device 1 determines similarity between the plurality of terms on the basis of the generated feature vector 23 for each term.SELECTED DRAWING: Figure 1
【課題】同義語または多義語の抽出精度を向上する。【解決手段】情報処理装置1は、複数の文章と複数の文章それぞれに含まれる用語の指定とを受け付け、受け付けた複数の文章それぞれを意味解析して、指定を受け付けた用語ごとに、該用語と、該用語を含む文章に含まれる他の複数の用語それぞれとの意味構造22を生成する。情報処理装置1は、指定を受け付けた用語ごとに、生成した意味構造22から特定の意味構造22を抽出し、抽出した意味構造22を素性とする特徴ベクトル23を生成する。情報処理装置1は、生成した用語ごとの特徴ベクトル23に基づき、複数の用語間の類似性に関する判定を行う。【選択図】図1 |
---|