An extraction method of product feature keywords from press releases using syntactic dependency analysis with an expression phrase

被引：0
|
作者：
Samejima, Masaki [1 ,2 ,3 ]
Shimizu, Yutaka [1 ,3 ]
Akiyoshi, Masanori [1 ,3 ]
Komoda, Norihisa [1 ,3 ]
机构：
[1] Osaka University, Graduate School of Information Science and Technology, 2-1, Yamadaoka, Suita, Osaka 565-0871, Japan
[2] Hitachi, Ltd., System Development Laboratory, 292 Yoshida-cho, Yokohama, Kanagawa 244-0817, Japan
[3] Osaka University, Japan
来源：
IEEJ Transactions on Electronics, Information and Systems | 2007年 / 127卷 / 09期
关键词：
Industrial economics - Social aspects - Syntactics;
D O I：
10.1541/ieejeiss.127.1436
中图分类号：
学科分类号：
摘要：
We propose an extraction method of product features from press releases of companies for competitor analysis. In press releases of companies, product features are described with characteristic expressions (we call description patterns). From this view point, when a description pattern is input, our system extracts features included in the pattern. Additionally, in order to extract more product features by less input patterns, our system generates similar patterns of input patterns as derivation patterns automatically and extracts product features using derivation patterns. Through application to the press releases of ten companies, it is confirmed that the accuracy of the extracted features are sufficient.
引用
收藏
页码：1436 / 1441
相关论文
共 50 条
  • [31] A method of stall recognition using nonlinear feature extraction from the compressor outlet pressure
    Mlkvik, Marek
    Olsiak, Robert
    Knizat, Branislav
    HELIYON, 2023, 9 (10)
  • [32] Reconstructive discriminant analysis: A feature extraction method induced from linear regression classification
    Chen, Yi
    Jin, Zhong
    NEUROCOMPUTING, 2012, 87 : 41 - 50
  • [33] Multimodal independent component analysis - A method of feature extraction from multiple information sources
    Akaho, S
    Umeyama, S
    ELECTRONICS AND COMMUNICATIONS IN JAPAN PART III-FUNDAMENTAL ELECTRONIC SCIENCE, 2001, 84 (11): : 21 - 28
  • [34] Editorial Expression of Concern: Traffic sign detection and recognition using RGSM and a novel feature extraction method
    M. Sudha
    V. P. Galdis pushparathi
    Peer-to-Peer Networking and Applications, 2023, 16 (3) : 1572 - 1572
  • [35] A Fusion Feature Extraction Method Using EEMD and Correlation Coefficient Analysis for Bearing Fault Diagnosis
    Jiang, Fan
    Zhu, Zhencai
    Li, Wei
    Ren, Yong
    Zhou, Gongbo
    Chang, Yonggen
    APPLIED SCIENCES-BASEL, 2018, 8 (09):
  • [36] A Novel Method for Feature Extraction and Automatic Recognition of Tire Defects Using Independent Component Analysis
    Cui, Xue Hong
    Liu, Yun
    Wang, Chuan Xu
    Li, Hui
    INTERNATIONAL CONFERENCE ON INTELLIGENT MANUFACTURING AND MATERIALS, 2016, 2016, : 141 - 149
  • [37] A data processing method based on sequence labeling and syntactic analysis for extracting new sentiment words from product reviews
    Zhang, Shunxiang
    Xu, Hanqing
    Zhu, Guangli
    Chen, Xiang
    Li, KuanChing
    SOFT COMPUTING, 2022, 26 (02) : 853 - 866
  • [38] A data processing method based on sequence labeling and syntactic analysis for extracting new sentiment words from product reviews
    Shunxiang Zhang
    Hanqing Xu
    Guangli Zhu
    Xiang Chen
    KuanChing Li
    Soft Computing, 2022, 26 : 853 - 866
  • [39] RCS feature extraction from simple targets using time-frequency analysis
    Rasmussen, JL
    Haupt, RL
    Walker, MJ
    RADAR PROCESSING, TECHNOLOGY, AND APPLICATIONS, 1996, 2845 : 66 - 74
  • [40] Facial feature extraction from a video sequence using Independent Component Analysis (ICA)
    Choi, KY
    Takaya, K
    2001 IEEE PACIFIC RIM CONFERENCE ON COMMUNICATIONS, COMPUTERS AND SIGNAL PROCESSING, VOLS I AND II, CONFERENCE PROCEEDINGS, 2001, : 259 - 262
12345