WORD EXTRACTION DEVICE, WORD EXTRACTION SYSTEM, AND WORD EXTRACTION METHOD

To improve the accuracy of word extraction.SOLUTION: A word extraction device includes: a sentence expression generation part for acquiring teacher data including a sentence with a word of an extraction object specified, generating a first sentence expression by processing the teacher data by a firs...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: HONMA TAKESHI, EKANT MULJIBHAI AMIN
Format: Patent
Sprache:eng ; jpn
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:To improve the accuracy of word extraction.SOLUTION: A word extraction device includes: a sentence expression generation part for acquiring teacher data including a sentence with a word of an extraction object specified, generating a first sentence expression by processing the teacher data by a first sentence analysis method, generating a second sentence expression by processing the teacher data by a second analysis method, and generating a first synthetic sentence expression by synthesizing the first sentence expression with the second sentence expression; a query expression generation part for generating an extraction query expression showing a query for extracting the word of an extraction object from a prescribed retrieval object document on the basis of the first synthetic sentence expression; and a word extraction part for extracting extraction information showing information related to the word of an extraction object from a second synthetic sentence expression to be generated on the basis of the retrieval object document by using the extraction query expression.SELECTED DRAWING: Figure 4 【課題】単語抽出の精度を向上させること。【解決手段】抽出対象の単語が特定されている文章を含む教師データを取得し、教師データを第1の文解析手法によって処理することで第1の文章表現を生成し、教師データを第2の文解析手法によって処理することで第2の文章表現を生成し、第1の文章表現と、第2の文章表現とを合成することで第1の合成文章表現を生成する文章表現生成部と、第1の合成文章表現に基づいて、抽出対象の単語を所定の検索対象文献から抽出するクエリを示す抽出クエリ表現を生成するクエリ表現生成部と、抽出クエリ表現を用いることで、抽出対象の単語に関する情報を示す抽出情報を検索対象文献に基づいて生成される第2の合成文章表現から抽出する単語抽出部とを含む単語抽出装置。【選択図】図4