應用興趣點辨識技術從 Web 中挖掘新商家資訊

Kuo Hsin Hsu; Hsiu Min Chuang; Chien Lung Chou; Chia Hui Chang

應用興趣點辨識技術從 Web 中挖掘新商家資訊

Translated title of the contribution: Mining POIs from web via POI recognition and relation verification

Kuo Hsin Hsu, Hsiu Min Chuang, Chien Lung Chou, Chia Hui Chang

Department of Computer Science and Information Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

This paper presents a system that could automatically extract new POIs from Web. First, we use special queries (e.g. Taipei+New Open) to find Web pages that might contain addresses for new stores. For web pages that contain addresses, we then apply store name recognition model to extract possible POIs. Finally, we train a model to find the most possible POI for the address found in the page. In this paper, we focus on POI name recognition and POI relation prediction. For POI recognition, we use store names from yellow pages as seed to prepare the training data via distant learning. Through entity selection and data processing, we obtain a model with 0.816 F1-measure as opposed to 0.432 F1-measure for a dictionary-based baseline. As for POI relation prediction, we compare three different strategies for negative example preparation. The best model could get 0.754 accuracy. We combine two POI recognition models with three classification models to test the overall performance. The best combination could extract 49 POIs every day with a single IP.

Translated title of the contribution	Mining POIs from web via POI recognition and relation verification
Original language	Chinese (Traditional)
Title of host publication	Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017
Editors	Lun-Wei Ku, Yu Tsao, Chi-Chun Lee, Cheng-Zen Yang, Hung-Yi Lee, Richard T.-H. Tsai, Wen-Hsiang Lu, Shih-Hung Wu
Publisher	The Association for Computational Linguistics and Chinese Language Processing (ACLCLP)
Pages	53-67
Number of pages	15
ISBN (Electronic)	9789869576901
State	Published - 1 Nov 2017
Event	29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017 - Taipei, Taiwan Duration: 27 Nov 2017 → 28 Nov 2017

Publication series

Name	Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017

Conference

Conference	29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017
Country/Territory	Taiwan
City	Taipei
Period	27/11/17 → 28/11/17

Cite this

Hsu, K. H., Chuang, H. M., Chou, C. L., & Chang, C. H. (2017). 應用興趣點辨識技術從 Web 中挖掘新商家資訊. In L-W. Ku, Y. Tsao, C-C. Lee, C-Z. Yang, H-Y. Lee, R. T-H. Tsai, W-H. Lu, & S-H. Wu (Eds.), Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017 (pp. 53-67). (Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017). The Association for Computational Linguistics and Chinese Language Processing (ACLCLP).

Hsu, Kuo Hsin ; Chuang, Hsiu Min ; Chou, Chien Lung et al. / 應用興趣點辨識技術從 Web 中挖掘新商家資訊. Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017. editor / Lun-Wei Ku ; Yu Tsao ; Chi-Chun Lee ; Cheng-Zen Yang ; Hung-Yi Lee ; Richard T.-H. Tsai ; Wen-Hsiang Lu ; Shih-Hung Wu. The Association for Computational Linguistics and Chinese Language Processing (ACLCLP), 2017. pp. 53-67 (Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017).

@inproceedings{6e0393b0e0d142deb772e96fe45154ab,

title = "應用興趣點辨識技術從 Web 中挖掘新商家資訊",

abstract = "This paper presents a system that could automatically extract new POIs from Web. First, we use special queries (e.g. Taipei+New Open) to find Web pages that might contain addresses for new stores. For web pages that contain addresses, we then apply store name recognition model to extract possible POIs. Finally, we train a model to find the most possible POI for the address found in the page. In this paper, we focus on POI name recognition and POI relation prediction. For POI recognition, we use store names from yellow pages as seed to prepare the training data via distant learning. Through entity selection and data processing, we obtain a model with 0.816 F1-measure as opposed to 0.432 F1-measure for a dictionary-based baseline. As for POI relation prediction, we compare three different strategies for negative example preparation. The best model could get 0.754 accuracy. We combine two POI recognition models with three classification models to test the overall performance. The best combination could extract 49 POIs every day with a single IP.",

keywords = "Address Recognition, POI Entity Recognition, POI Relation Prediction",

author = "Hsu, {Kuo Hsin} and Chuang, {Hsiu Min} and Chou, {Chien Lung} and Chang, {Chia Hui}",

note = "Publisher Copyright: {\textcopyright} The Association for Computational Linguistics and Chinese Language Processing; 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017 ; Conference date: 27-11-2017 Through 28-11-2017",

year = "2017",

month = nov,

day = "1",

language = "繁體中文",

series = "Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017",

publisher = "The Association for Computational Linguistics and Chinese Language Processing (ACLCLP)",

pages = "53--67",

editor = "Lun-Wei Ku and Yu Tsao and Chi-Chun Lee and Cheng-Zen Yang and Hung-Yi Lee and Tsai, {Richard T.-H.} and Wen-Hsiang Lu and Shih-Hung Wu",

booktitle = "Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017",

}

Hsu, KH, Chuang, HM, Chou, CL & Chang, CH 2017, 應用興趣點辨識技術從 Web 中挖掘新商家資訊. in L-W Ku, Y Tsao, C-C Lee, C-Z Yang, H-Y Lee, RT-H Tsai, W-H Lu & S-H Wu (eds), Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017. Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017, The Association for Computational Linguistics and Chinese Language Processing (ACLCLP), pp. 53-67, 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017, Taipei, Taiwan, 27/11/17.

應用興趣點辨識技術從 Web 中挖掘新商家資訊. / Hsu, Kuo Hsin; Chuang, Hsiu Min; Chou, Chien Lung et al.
Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017. ed. / Lun-Wei Ku; Yu Tsao; Chi-Chun Lee; Cheng-Zen Yang; Hung-Yi Lee; Richard T.-H. Tsai; Wen-Hsiang Lu; Shih-Hung Wu. The Association for Computational Linguistics and Chinese Language Processing (ACLCLP), 2017. p. 53-67 (Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - 應用興趣點辨識技術從 Web 中挖掘新商家資訊

AU - Hsu, Kuo Hsin

AU - Chuang, Hsiu Min

AU - Chou, Chien Lung

AU - Chang, Chia Hui

N1 - Publisher Copyright: © The Association for Computational Linguistics and Chinese Language Processing

PY - 2017/11/1

Y1 - 2017/11/1

N2 - This paper presents a system that could automatically extract new POIs from Web. First, we use special queries (e.g. Taipei+New Open) to find Web pages that might contain addresses for new stores. For web pages that contain addresses, we then apply store name recognition model to extract possible POIs. Finally, we train a model to find the most possible POI for the address found in the page. In this paper, we focus on POI name recognition and POI relation prediction. For POI recognition, we use store names from yellow pages as seed to prepare the training data via distant learning. Through entity selection and data processing, we obtain a model with 0.816 F1-measure as opposed to 0.432 F1-measure for a dictionary-based baseline. As for POI relation prediction, we compare three different strategies for negative example preparation. The best model could get 0.754 accuracy. We combine two POI recognition models with three classification models to test the overall performance. The best combination could extract 49 POIs every day with a single IP.

AB - This paper presents a system that could automatically extract new POIs from Web. First, we use special queries (e.g. Taipei+New Open) to find Web pages that might contain addresses for new stores. For web pages that contain addresses, we then apply store name recognition model to extract possible POIs. Finally, we train a model to find the most possible POI for the address found in the page. In this paper, we focus on POI name recognition and POI relation prediction. For POI recognition, we use store names from yellow pages as seed to prepare the training data via distant learning. Through entity selection and data processing, we obtain a model with 0.816 F1-measure as opposed to 0.432 F1-measure for a dictionary-based baseline. As for POI relation prediction, we compare three different strategies for negative example preparation. The best model could get 0.754 accuracy. We combine two POI recognition models with three classification models to test the overall performance. The best combination could extract 49 POIs every day with a single IP.

KW - Address Recognition

KW - POI Entity Recognition

KW - POI Relation Prediction

UR - http://www.scopus.com/inward/record.url?scp=85085935032&partnerID=8YFLogxK

M3 - 會議論文篇章

AN - SCOPUS:85085935032

T3 - Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017

SP - 53

EP - 67

BT - Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017

A2 - Ku, Lun-Wei

A2 - Tsao, Yu

A2 - Lee, Chi-Chun

A2 - Yang, Cheng-Zen

A2 - Lee, Hung-Yi

A2 - Tsai, Richard T.-H.

A2 - Lu, Wen-Hsiang

A2 - Wu, Shih-Hung

PB - The Association for Computational Linguistics and Chinese Language Processing (ACLCLP)

T2 - 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017

Y2 - 27 November 2017 through 28 November 2017

ER -

Hsu KH, Chuang HM, Chou CL, Chang CH. 應用興趣點辨識技術從 Web 中挖掘新商家資訊. In Ku L-W, Tsao Y, Lee C-C, Yang C-Z, Lee H-Y, Tsai RT-H, Lu W-H, Wu S-H, editors, Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017. The Association for Computational Linguistics and Chinese Language Processing (ACLCLP). 2017. p. 53-67. (Proceedings of the 29th Conference on Computational Linguistics and Speech Processing, ROCLING 2017).

應用興趣點辨識技術從 Web 中挖掘新商家資訊

Abstract

Publication series

Conference

Fingerprint

Cite this