Community Question Answering Forums Research Articles

Feature engineering in learning-to-rank for community question answering task

Community question answering (CQA) forums are Internet-based platforms where users ask questions about a topic and other expert users try to provide solutions. Many CQA forums such as Quora, Stackoverflow, Yahoo!Answer, StackExchange exist with a lot of user-generated data. These data are leveraged in automated CQA ranking systems where similar questions (and answers) are presented in response to the user's query. In this work, we empirically investigate a few aspects of this domain. First, in addition to traditional features like Term Frequency (TF), Inverse Document Frequency (IDF), Best Match 25 (BM25), etc., we introduce a Bidirectional Encoder Representations from Transformers (BERT)-based feature that captures the semantic similarity between the question and answer. Second, most of the existing research works have focus on features extracted only from the question part; features extracted from answers have not been explored extensively. We combine both types of features in a linear fashion. Third, using our proposed concepts, we conduct an empirical investigation with different rank-learning algorithms, some of which have not been used so far in CQA domain. On three standard CQA datasets, our proposed framework achieves state-of-the-art performance (0.56, 0.58 and 0.60 NDCG@10 values). We also analyze the importance of the features we use in our investigation. This work is expected to guide the practitioners to devise a better set of features for the CQA retrieval task.

A Study on Influential Features for Predicting Best Answers in Community Question-Answering Forums

The knowledge provided by user communities in question-answering (QA) forums is a highly valuable source of information for satisfying user information needs. However, finding the best answer for a posted question can be challenging. User-generated content in forums can be of unequal quality given the free nature of natural language and the varied levels of user expertise. Answers to a question posted in a forum are compiled in a discussion thread, concentrating also posterior activity such as comments and votes. There are usually multiple reasons why an answer successfully fulfills a certain information need and gets accepted as the best answer among a (possibly) high number of answers. In this work, we study the influence that different aspects of answers have on the prediction of the best answers in a QA forum. We collected the discussion threads of a real-world forum concerning computer programming, and we evaluated different features for representing the answers and the context in which they appear in a thread. Multiple classification models were used to compare the performance of the different features, finding that readability is one of the most important factors for detecting the best answers. The goal of this study is to shed some light on the reasons why answers are more likely to receive more votes and be selected as the best answer for a posted question. Such knowledge enables users to enhance their answers which leads, in turn, to an improvement in the overall quality of the content produced in a platform.

Retracted: Combining Latent Factor Model for Dynamic Recommendations in Community Question Answering Forums

Retracted: Combining Latent Factor Model for Dynamic Recommendations in Community Question Answering Forums

Social brokerage and productivity of users in online innovation networks

AbstractIn this work, we investigate the impact of social brokerage on the innovation productivity of users in online innovation networks. Innovation in online networks is leveraged by coordinated interactions between the participating users. We leverage the data of users contributing on posts between 2017 and 2019 in two community question answering (CQA) forums: Stack Overflow and Math Stack Exchange. The innovation productivity of users in such online innovation networks, as valued by other participating users is quantified by the total score they received on their contributing posts. Users occupying brokerage positions are exposed to better ideas that boost the value of their contributing posts, thus improving their total score. Again, existing users with low social brokerage are also likely to gain from resource sharing owing to their denser networks. Our econometric analysis reveals that users in online innovation networks with very low or very high values of social brokerage receive higher scores as valued by other community users. Overall, we access the data of 537,938 users in Stack Overflow contributing on 6,002,996 questions, and 7,903,416 answers, and 20,393 users in the Math Stack Exchange contributing on 488,389 questions, and 638,110 answers. We observe a U‐shaped effect of the user's social brokerage on the user's innovation productivity. Our empirical findings have various implications for firms hosting online CQA forums.

Combining Latent Factor Model for Dynamic Recommendations in Community Question Answering Forums.

Community Question Answering (CQA) web service provides a platform for people to share knowledge. Quora, Stack Overflow, and Yahoo! Answers are few sites where questioners post their queries and answerers respond to their respective queries. Due to the ease of use and quick responsiveness of the CQA platform, these sites are being widely adopted by the community. For better usability, there is a dire need to route the question toward the relevant answerers. To fulfil this gap, recommender systems play an important role in identifying the relevant answerers. To map the user interests more effectively, this research work proposed a dynamic feature representation of the latent user attributes for user profiling. The latent features are mapped by leveraging the Latent Dirichlet Allocation (LDA) for topic modelling of user data. The proposed recommendation model segments the user profile based on these latent user profiles incorporating the incremental learning of the users' interests to produce the relevant recommendations in near real time. The experimental setup generated recommendation lists of variable sizes and evaluated using multiple evaluation metrics, such as mean average precision, recall, throughput, and different quality metrics, such as discounted cumulative gain and mean reciprocal rank. The results showed that the proposed model provided a better quality of recommendations in CQA forums, which is promising for future research in this domain.

Informative Multimedia QA using Web based Approach

question answering (QA) still faces challenges such as processing and deep understanding of complex questions. In some cases, human intelligence obtains better results than automated approach. The result shows that community question answering (CQA) emerged as an extremely popular alternative to obtain information in which users are able to obtain better answers provided by other participants. But existing CQA forums mostly support textual answers, which are not informative enough for many questions. In this paper we propose a system that enriches textual answers with corresponding media data in Community QA. Our model consists of three components; query analysis for multimedia search, answer medium selection and multimedia data selection and presentation. This system automatically determines which type of media information should be added for textual answer by collecting data from web to enrich the answer. Multimedia QA scheme uses diversification methods to collect the best suitable answers based on questions and make the enriched media data more diverse. It determines the type of medium to be used by adding Nave Bayes Classifier, which helps to generate queries based on existing QA dataset pool and performs multimedia search. This scheme also performs query adaptive re-ranking and redundancy removal to obtain a set of images and videos for presentation accompanying textual answer. It uses Page Ranking algorithm which result shows that it provides more satisfactory and effective results Keywordsmedium selection, question answering, Reranking

Question Quality in Community Question Answering Forums

Community Question Answering websites (CQA) offer a new opportunity for users to provide, search and share knowledge. Although the idea of receiving a direct, targeted response to a question sounds very attractive, the quality of the question itself can have an important effect on the likelihood of getting useful answers. High quality questions improve the CQA experience and therefore it is essential for CQA forums to better understand what characterizes questions that are more appealing for the forum community. In this survey, we review existing research on question quality in CQA websites. We discuss the possible measures of question quality and the question features that have been shown to influence question quality.