Joint Model of Topics, Expertises, Activities and Trends for Question Answering Web Applications

Abstract

HAL is a multidisciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

IEA: an answerer recommendation approach on stack overflow

“…These existing approaches consider user topical interest and expertise to recommend answerers. Other previous work [4] analyzed user activeness to recommend answerers in CQA websites. In CQA websites, users not only post questions and answers in CQA websites, but also post comments on these questions and answers.…”

“…In CQA websites, users not only post questions and answers in CQA websites, but also post comments on these questions and answers. In existing work [4], user activeness is analyzed based on historical questions and answers, without considering comments. Besides, existing answerer recommendation approaches only consider user topical interest and expertise [3,4], or user topical interest and activeness [4].…”

“…In existing work [4], user activeness is analyzed based on historical questions and answers, without considering comments. Besides, existing answerer recommendation approaches only consider user topical interest and expertise [3,4], or user topical interest and activeness [4]. However, they do not combine topical interest, topical expertise and activeness together to recommend answerers.…”

“…Specifically, IEA achieves nDCG@1, nDCG@5, nDCG, Pearson rank correlation coefficient and Kendall rank correlation coefficient values of 0.6624, 0.8349, 0.9020, 0.1880 and 0.1649, respectively. Besides, in comparison with the approaches of TEM [3], temporal topic expertise activity (TTEA) [4] and TTEA-activeness level (TTEA-ACT) [4], IEA improves nDCG@1 by 10.29%, 14.53% and 15.17%, and improves nDCG@5 by 2.68%, 3.74% and 4.11%, and improves nDCG by 2.48%, 3.45% and 3.79%, and improves Pearson rank correlation coefficient by 236.20%, 84.91% and 224.12%, and improves Kendall rank correlation coefficient by 424.18%, 1845.30% and 772.60%. Besides, we analyze the effect of the number of topics and expertise to answerer recommendation.…”

“…• Results show that IEA outperforms TEM [3], TTEA [4] and TTEA-ACT [4] by substantial margins. The reminder of the paper is organized as follows.…”