Knowledge Demand and Supply Analysis in Community Questions Answering Based on TF-PIDF
Li Ming,Li Ying,Zhou Qing,Wang Jun
(School of Economics and Management, China University of Petroleum-Beijing, Beijing 102249, China)
(School of Economics and Management, Beihang University, Beijing 100191, China)
[Objective] Community Question Answering (CQA) is an important platform for knowledge sharing. In order to master people's knowledge demand and knowledge supply in CQA, and then make targeted interventions, in the paper, the method for analyzing knowledge demand and knowledge supply in community questions answering is proposed.
[Methods] Firstly, in view of the characteristics that the questions and answers in the question and answer pair are composed of multiple parts, a novel word weight calculation method TF-PIDF is constructed to model the questions and answers respectively. The main categories of knowledge demand and knowledge supply are obtained by clustering the questions and answers, and then the topics and hot degree of each category are obtained. For each category of knowledge demand, the main aspects of the corresponding knowledge supply are identified. The knowledge demand coverage algorithm is proposed to calculate the coverage of knowledge demand by knowledge supply, then it is proposed to carry out a cross-analysis of knowledge demand from heat and coverage.
[Results] The case study on influenza in the community ZHIHU is conducted. Six topic categories of knowledge demand and knowledge supply are obtained, respectively. The hot topic is both "epidemic", but its coverage of knowledge supply is low, which is considered to be the hot real-time knowledge demand under the emergency of influenza. The experimental results show that the method is feasible and performs well.
[Limitations] Although the analysis framework and method proposed in this paper can effectively mine the topic of knowledge demand and knowledge supply in CQA, the identified topic still stays on the topic meaning level expressed by the feature word clustering.
[Conclusions] With the proposed method, the knowledge demand of people and knowledge supply of CQA are mastered. It also provides important basis for knowledge supplement and operation of the community.
李明, 李莹, 周庆, 王君.
基于TF-PIDF的网络问答社区中的知识供需研究
[J]. 数据分析与知识发现, 10.11925/infotech.2096-3467. 2020.0395.
Li Ming, Li Ying, Zhou Qing, Wang Jun.
Knowledge Demand and Supply Analysis in Community Questions Answering Based on TF-PIDF
. Data Analysis and Knowledge Discovery, 0, (): 1-.