Text Category Crowdsourcing Solution Filter Research in Reward Mode

DOI

10.2991/mmetss-16.2017.22How to use a DOI?

Keywords

crowdsourcing, filter, reward mode, solution selection, double-layer filter.

Abstract

A two-level filtering model is proposed to solve the problem of crowdsourcing scheme selection. Firstly, we use word segmentation tool NLPIR to segment crowdsourcing word, then get the text key words and word frequency statistics. Secondly, according to the word frequency method, we extract feature words and establish feature matrix. Thirdly, the vector space model is used to describe the text contents of the crowdsourcing, and the double layer filter is built according to the text relevance theory. Finally, To verify validity of the filtering model, we use it to filter the existing historical crowdsourcing schemes on the crowdsourcing platform, and compare the results of filtering with the results of artificial selection. It shows that this filtering model can realize the semi-automatic selection of crowdsourcing scheme. It can filtrate a large number of ineffective schemes and keep effective schemes. The two-level filtering model has practical application value.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Huixing Nie
AU  - Tianshun Wang
PY  - 2017/02
DA  - 2017/02
TI  - Text Category Crowdsourcing Solution Filter Research in Reward Mode
BT  - Proceedings of the 2016 International Conference on Modern Management, Education Technology, and Social Science (MMETSS 2016)
PB  - Atlantis Press
SP  - 117
EP  - 124
SN  - 2352-5398
UR  - https://doi.org/10.2991/mmetss-16.2017.22
DO  - 10.2991/mmetss-16.2017.22
ID  - Nie2017/02
ER  -