Answering Communities Research Articles

Visual potential expert prediction in question and answering communities

The success of Question and Answering (Q&A) communities mainly depends on the contribution of experts. However, there is a bottleneck for machine to identify these experts as soon as they participate in a community due to lack of enough activities during users’ early participation. To tackle that, we bring human’s business experience to potential expert prediction by combining machine learning and visual analytics. In this work, we propose a visual analytics system to identify potential experts semi-automatically. After the machine learning algorithm gives the result of the expert probability, analysts can locate a set of interested users whose expert probability is ambiguous and check the user information and behavior patterns of those users via the design of multi-dimension data visualization. Finally, our system models analysts’ knowledge of the community members’ identities, and then abstracts the knowledge quantificationally for machine learning algorithm. Thus, analysts can modify machine learning algorithm and the prediction process smoothly. A quantitative evaluation with real data has been studied to demonstrate the effectiveness of our system.

A survey on mining stack overflow: question and answering (Q&A) community

PurposeSoftware developers extensively use stack overflow (SO) for knowledge sharing on software development. Thus, software engineering researchers have started mining the structured/unstructured data present in certain software repositories including the Q&A software developer community SO, with the aim to improve software development. The purpose of this paper is show that how academics/practitioners can get benefit from the valuable user-generated content shared on various online social networks, specifically from Q&A community SO for software development.Design/methodology/approachA comprehensive literature review was conducted and 166 research papers on SO were categorized about software development from the inception of SO till June 2016.FindingsMost of the studies revolve around a limited number of software development tasks; approximately 70 percent of the papers used millions of posts data, applied basic machine learning methods, and conducted investigations semi-automatically and quantitative studies. Thus, future research should focus on the overcoming existing identified challenges and gaps.Practical implicationsThe work on SO is classified into two main categories; “SO design and usage” and “SO content applications.” These categories not only give insights to Q&A forum providers about the shortcomings in design and usage of such forums but also provide ways to overcome them in future. It also enables software developers to exploit such forums for the identified under-utilized tasks of software development.Originality/valueThe study is the first of its kind to explore the work on SO about software development and makes an original contribution by presenting a comprehensive review, design/usage shortcomings of Q&A sites, and future research challenges.

An Algorithm Research on Potential Problem Reactor in Medical Online Question Answering Community

Objective: Based on the online question-and-answer community of Internet medicine, a weighted user-personal model is proposed in viewof the pain point of the users' queries in the medical question-and-answer community which can not be effectively answered in time. Predictswhether the recommended user can respond timely and effectively to the problem to improve the accuracy of the proposed results. .Methods: Using the user's historical activity data in the community, a model of user's personal domain was proposed based on query-likelanguage model to describe the degree of user's concern in different field.Results: In the experiment based on Baidu medical question-answering system data, compared with the existing algorithms, the potentialproblem respondents in this paper have better recommendation effect.

Discovering Top Experts for Trending Domains on Stack Overflow

Stack Overflow is an online Question Answer Community, generating huge user content for research analysis. Stack Overflow is used to find top experts on a specific domain. The domains of interest are java, javascript, C# and python. The study identifies topic specific expert as with the Stack Overflow statistics, where the latter provides top experts based on reputation score irrespective of the domain. The algorithms used are PageRank and HITS score. The Discounted Cumulative Gain is used to measure the correctness of the analysis.

Beyond reciprocity: The bystander effect of knowledge response in online knowledge communities

Because new members are important sources of knowledge to online knowledge communities, it is important to retain them after their initial interactions with the community. With a large-scale behavioral dataset collected from a leading online Question and Answer community for programmers, Stack Overflow, we investigated how the community's knowledge responses and social responses to newcomers' questions affected their subsequent likelihood of knowledge contribution (answering others' questions) and knowledge seeking (asking more questions). Contrary to the theory of reciprocity, and in line with predictions by the bystander effect, we found that receiving high quality answers negatively influenced new knowledge seekers' future likelihood of knowledge contribution. Consistent with the social exchange theory, receiving high quality answers positively affected newcomers' future knowledge seeking behaviors. Social responses (votes to the new members' questions) were found to have strong positive effects on both newcomers' future knowledge contribution and seeking behaviors.

A hybrid model to predict best answers in question answering communities

Question answering communities (QAC) are nowadays becoming widely used due to the huge facilities and flow of information that it provides. These communities target is to share and exchange knowledge between users. Through asking and answering questions under large number of categories.Unfortunately there are a lot of issues existing that made knowledge process difficult. One of those issues is that not every asker has the knowledge and ability to select the best answer for his question, or even selecting the best answer based on subjective matters. Our analysis in this paper is conducted on stack overflow community. We proposed a hybrid model for predicting the best answer. The proposed model is consisting of two modules. The first module is the content feature which consists of three types of features; question-answer features, answer content features, and answer-answer features. In the second module we examine the use of non-content feature in predicting best answers by using novel reputation score function. Then we merge both of content and non-content features and use them in prediction. We conducted experiments to train three different classifiers using our new added features. The prediction accuracy is very promising.

Separating the Wheat from the Chaff: Evaluating Success Determinants for Online Q&A Communities

Researchers and community managers try to measure the success of online communities using a variety of success determinants, such as member activity, turnover and interaction. Although many success determinants have been proposed in the literature, it remains largely unclear which are applicable to which kind of online communities. In this work, we focus on the use case of question answering (Q&A) communities, where we can measure asker satisfaction as an aspect of community success towards the goal of solving questions. We evaluate how well the proposed success determinants correlate to asker satisfaction, and we find that less than half of them are at least moderately (±0.3 or more) correlated. That underlines the importance of evaluating the proposed success determinants on real data of specific community types.

Beyond Reciprocity: The Bystander Effect of Knowledge Response in Online Knowledge Communities

Because new members are important sources of knowledge to online knowledge communities, it is important to retain them after their initial interactions with the community. With a large-scale behavioral dataset collected from a leading online Question and Answer community for programmers, Stack Overflow, we investigated how the community's knowledge responses and social responses to newcomers' questions affected their subsequent likelihood of knowledge contribution (answering others' questions) and knowledge seeking (asking more questions). Contrary to the theory of reciprocity, and in line with predictions by the bystander effect, we found that receiving high quality answers negatively influenced new knowledge seekers' future likelihood of knowledge contribution. Consistent with the social exchange theory, receiving high quality answers positively affected newcomers' future knowledge seeking behaviors. Social responses (votes to the new members' questions) were found to have strong positive effects on both newcomers' future knowledge contribution and seeking behaviors.

Belief Measure of Expertise for Experts Detection in Question Answering Communities: case study Stack Overflow

Online Question Answering Communities (Q& A C) provide a valuable amount of information in several topics. The major challenge with Q& A C is the detection of the authoritative users. When manipulating real world data, we have to deal with imperfections and uncertainty that can occur. In this paper, we propose a belief measure of expertise allowing us to detect users with the highest degree of expertise based on their attributes. Experiments on a dataset from a large online Q&A Community prove that the proposed model can be used to improve the identification of most expert users.

Does the Framing of Progress Towards Virtual Rewards Matter?

A natural experiment on a popular German Question & Answer community is used to find out whether the small-area hypothesis applies to user activation by means of a virtual reward in the form of badges. Koo and Fishbach’s small-area hypothesis posits that individuals in pursuit of a goal are more highly motivated when focusing on the smaller percentage of progress towards their goal, irrespective of whether this figure represents the actions already completed or those still remaining. Consistent with the authors’ theoretical predictions, the study finds empirical evidence for the small-area effect and its activating power, translated here into increased online user contributions. Besides contributing to the literature with an empirical study anchored in theory, the findings have direct practical implications for designers of online virtual reward systems by suggesting more effective (and motivating) ways of framing user progress towards virtual rewards.

Question Recommendation and Answer Extraction in Question Answering Community

Every day, there are a large number of new questions produce in question answering community, how to find the answer user for the question and sort candidate answers is the research content of this paper. First of all ,we use statistical language model to model for user interest, make full use of the abundant personalized information in question answering community to find out user interest distribution, and obtain the user list of question recommendation by introducing the query likelihood language model to calculate the degree of user interest to the new question. Secondly, we calculate the matching degree of question and candidate answers through fusing the feature of word form, word order, distance and semantic. The candidate answers of question will be sorted automatically, making it easier for users to choose the best answer. Experiments are performed on data sets extracted from the Baidu know, experimental results show that the method proposed in this paper has better performance.

Using visual analytics to support online community development

Question and answer communities are an integral part of the web, enabling individuals to find answers to difficult technical questions while reducing the need for companies to host expensive technical support centres. Developing and maintaining these communities is a challenge given how online communities can scale considerably and are subject to continuous change. In this article, we present two case studies in which we investigate how visual analytics could be designed to support different online question and answer communities. In the first study, we explore how collaborative visual analytics can encourage analysis and conversation in the Super User online community while in the second study we investigate how visual analytics can be designed to support a strategic approach to online community management at Symantec. Our findings suggest that visualisation promotes discussion and reflection but does not readily produce actionable or repeatable insights. More effective solutions require a deeper understanding of community processes, enabling more predictive or actionable insights for different community members.

Mining Intention-Related Products on Online Q&A Community

User generated content on social media has attracted much attention from service/product providers, as it contains plenty of potential commercial opportunities. However, previous work mainly focuses on user consumption intention (CI) identification, and little effort has been spent to mine intention-related products. In this paper, focusing on the Baby & Child Care domain, we propose a novel approach to mine intention-related products on online question and answer (Q&A) community. Making use of the question-answering pairs as data source, we first automatically extract candidate products based on dependency parser. And then by means of the collocation extraction model, we identify the real intention-related products from the candidate set. The experimental results on our carefully constructed evaluation dataset show that our approach achieves better performance than two natural baseline methods.

Incorporating social information to perform diverse replier recommendation in question and answer communities

Social information is contextual information that has made significant contributions to intelligent information systems. However, social information has not been fully used, especially in question and answer (Q&A) systems. This study describes a contextual recommendation method in which diverse repliers are recommended for new questions using incorporated social information in Q&A communities. We have mined multiple kinds of social information by analysing social behaviours and relations found in a Q&A community and proposed an algorithm to incorporate different social information in various social contexts to perform diverse repliers’ recommendations. Recommendation diversity and social contexts have been considered and the properly used social information has been emphasized in this study. We conducted experiments using a dataset collected from the Stack Overflow website. The results demonstrate that different social information makes different contributions in promoting question answering, and incorporating social information properly could improve recommendation diversity and performance, which would then result in the promotion of satisfactory question solving.

Goals as Reference Points: Empirical Evidence from a Virtual Reward System

Heath et al. [Heath C, Larrick RP, Wu G (1999) Goals as reference points. Cognitive Psych. 38(1):79–109] propose a prospect theory model for goal behavior. Their analytical model is based on the assumption that goals inherit the main properties of the prospect theory value function, i.e., reference point dependence, loss aversion, and diminishing sensitivity. We investigate whether these main properties transfer to goal behavior in the field. We take user activity data from a gamified question and answer (Q&A) community and analyze how users adjust their contribution behavior in the days surrounding goal achievement, where goals are represented by badges. We find that users gradually increase their performance in the days prior to earning a badge, with performance peaking on the day of the promotion. In subsequent days, user performance gradually diminishes again, with the decline being strongest on the day immediately following the badge achievement. These findings reflect the characteristic S-shape of the prospect theory value function which is convex below the reference point and concave above it. Employing the target-based approach, we can interpret the value function as a cumulative density function of a unimodal probability distribution. Our results suggest that it is more likely that active members of the community focus on the next badge relative to the status already achieved, as their next goal and are less likely to focus on more remote (higher-ranked) badges. Our results thus support the transferability of the main properties of the prospect theory value function to goal behavior in the field and suggest a distinct shape of the value function around goals.

Inner reading voices: An overlooked form of inner speech

Inner speech is theorized to be the basis for auditory verbal hallucinations (AVHs), but few empirical studies have examined the phenomenology of inner speech, particularly while reading. One hundred and sixty posts from a popular question and answer community website were analyzed using a qualitative content analysis approach, to examine the phenomenology of inner reading voices (IRVs). Results indicated that many individuals report routinely experiencing IRVs, which often have the auditory qualities of overt speech, such as recognizable identity, gender, pitch, loudness and emotional tone. IRVs were sometimes identified as the readers’ own voices, and sometimes as the voices of other people. Some individuals reported that IRVs were continuous with audible thoughts. Both controllable and uncontrollable IRVs were reported. IRVs may provide evidence for individual variation in imagery vividness and support for inner speech accounts of AVHs. IRVs may be a useful model for studying AVHs in the non-clinical population and need further investigation.

Don’t take away my status! – Evidence from the restructuring of a virtual reward system

In a natural experiment on a popular German Question & Answer community we investigate how user status demotion affects user activity levels in online communities. The virtual reward system on this platform is designed to activate the status seeking behavior of its members. The members’ status within the community is represented by the member’s rank. In the experiment the platform operator restructured the virtual reward system, in the process reducing or abolishing the incentives for selected activities on the platform. After restructuring nearly all users saw their status demoted, with almost three quarters having lost one rank, and more than a quarter having lost two or more ranks. We identify the impact of status demotion by comparing how the members in these two groups respond to the restructuring. We find that users who lost two or more ranks reduce their post-event activities for which the incentives were reduced or abolished by 18% more than those who lost only one rank, and by 9% more than the latter for activities which were unaffected by the event.

An empirical study of topic-sensitive probabilistic model for expert finding in question answer communities

In this article, we study the problem of finding experts in community question answering (CQA). Most of the existing approaches attempt to find experts in CQA via link analysis. One primary challenge of expert finding lies in that how to improve authority score ranking based on the user information. However, these existing link analysis techniques largely fail to consider the interests, expertise, and reputation of users (question askers and answerers). To address this limitation, we propose a topic-sensitive probabilistic model, by extending the PageRank algorithm, more effectively find in the community by incorporating link and user analysis into a unified framework. We have conducted extensive experiments using a real world data set from Yahoo! Answers of English language. Results show that our method significantly outperforms the existing link analysis techniques and advances the state-of-the-art performance on expert finding in CQA.

Mining History of Gamification Towards Finding Expertise in Question and Answering Communities: Experience and Practice with Stack Exchange

Recently, an online Q&A tool has become an essential part of individual communities and organizations of experts on specific topics. Using the answers to questions about specific topics will help such communities work more efficiently in their fields. Currently, Q&A online communities are adopting gamification to engage users by granting awards to successful users. In this paper, we investigate how to mine award achievement histories to find expertise. We propose the use of sequence analysis and clustering techniques. Specifically, we study the history of Stack Exchange, a large Q&A community that employs gamification. To the best of our knowledge, this is the first study of using award achievement history to find expertise in Q&A communities.

An online partner for Holocaust remembrance education: students approaching the Yahoo! Answers community

Holocaust education has gained increased importance in recent decades and attention has latterly been directed to the role of the Internet within the field. Of major importance within the virtual space are Question and Answer communities. We investigated the interactions taking place within the Yahoo! Answers community following questions posted by students seeking help with their homework assignments, which required dealing with questions to ask a Holocaust survivor and writing an essay about Holocaust remembrance. The overall findings indicate that the majority of answerers responded willingly; however their answers are in most cases based upon their own views and reflect common notions, and are rarely accepted by askers. Yet several factors, among them the immense popularity of these communities, call for educators and scholars engaged in Holocaust remembrance education to develop closer acquaintance with this medium, which might play a considerable role in shaping the ways students approach Holocaust-related issues.