Question Answering Sites Research Articles

Examining user roles in social Q&A: The case of health topics in Zhihu.com

ABSTRACTSocial question answering (SQA) sites have become a popular online community platform for information seeking and social interaction. Prior SQA research has focused more on questions and answers, but has not examined user roles and their characteristics and activity patterns. In this poster, we identified four collaborative roles (starter, answerer, technical editor and follower) inZhihu.com, a popular expert SQA site in China, and examined their characteristics and activity patterns. Our data set contained 14,779 users, 2,362 questions, and 8,486 answers from 12 health‐related topics. Results show that, starters and technical editors ask more questions, make more technical edits, answers a few questions, and are not well connected to other users. Answerers contribute more content, attract the most followers and received the most likes and votes. Followers contribute little content, and receive the least likes and votes, but follow more topics and users. Content contributing activities such as answering questions gain more likes and votes, but posting a question and making a technical edit can get more pageviews. Users have overlapping roles and more research is needed to fully understand the interaction and interplay among user roles in SQA collaboration.

Folksonomy Based Trend Analysis on Community Question Answering Sites: A Perspective on Software Technologies

In the booming era of online social media environment, community question answering (CQA) sites have become one of the popular resources for software engineers and software industries. Software engineers are increasingly sharing their questions and answers on CQA sites. Aims of the CQA sites are to provide useful and relevant information to the users. Analysis information of major programming languages using trend analysis can be useful for software engineers to understand the technological evolutions and popularity. Since most of the CQA sites consist of user assigned tags by which folksonomy can be efficiently utilized for developing suitable algorithm to find the trend of key programming technologies. In this paper, two techniques of trend analysis, namely, ARIMA time series model and fuzzy time series model have been applied on the tagging data to identify the trend. Trend analysis is being carried on key programming languages, namely, c#, Java, PHP, and python. In this paper, quality of the trend is measured by entropy, $Z_{\mathrm {{Trend}}}$ , and quality of forecast is measured by MMRE, burst trend for key programming languages.

Search clicks analysis for discovering temporally anchored questions in community Question Answering

Nowadays, community Question-Answering (cQA) sites are massive repositories for user-generated content, where members prompt questions expecting satisfactory answers from other members. However, in this dynamic, there is an intrinsic delay between the moment questions are posted to the arrival of acceptable responses. Therefore, cQA platforms have the pressing need for promoting unresolved questions to potential answerers and for taking advantage of resolved questions contained in their archives, whenever possible.This paper studies cQA services from the viewpoint of the time frame where their questions attract the interest of their community members. By drawing a parallel with temporal patterns of user interests in web search activity, we are able to define three main types of temporally anchored questions: trend or bursty, periodic and permanent. Then, by analyzing user click distributions to Yahoo! Answers pages across Yahoo! Search logs, we automatically acquired a set of 35,000 cQA questions labeled with one of these three temporal anchors. Accordingly, we show the practicality of this approach by means of human assessments; and by using this automatically acquired corpus for studying several classification models.Essentially, the proposed method was found to correlate well with these human judgements, and proven to be effective in building systems that automatically identify the temporal anchor of unseen cQA questions. In substance, our outcomes indicate that some contexts are strongly related to a particular temporal anchor. We believe that these anchors will contribute to the discrimination of resolved questions that are capable of being revitalized, as well as to foster the opportune participation in questions that generate enthusiasm only for a short time.

Which Step Do I Take First? Troubleshooting with Bayesian Models

Online discussion forums and community question-answering websites provide one of the primary avenues for online users to share information. In this paper, we propose text mining techniques which aid users navigate troubleshooting-oriented data such as questions asked on forums and their suggested solutions. We introduce Bayesian generative models of the troubleshooting data and apply them to two interrelated tasks: (a) predicting the complexity of the solutions (e.g., plugging a keyboard in the computer is easier compared to installing a special driver) and (b) presenting them in a ranked order from least to most complex. Experimental results show that our models are on par with human performance on these tasks, while outperforming baselines based on solution length or readability.

User Motivation and the Effects of Multitasking: An Analysis of an Online CQA Forum

There is an increasing prevalence of Collaborative Question Answering (CQA) websites and a growing reliance of the online community on user-generated content. In this paper, we study users' motivation to win social recognition contests (best answers) and how multitasking and flexible hours influence the performance of contest-motivated users. Heterogeneity of contest motivation is estimated at the user level in a standard contest framework and used to demonstrate that for those who are highly motivated to win contests, multitasking across subjects is associated with lower productivity while flexible hours are associated with higher productivity, as proxied by the rate of winning contests.

Information Searching Using Web Data Extraction for Acquiring Enrich MM Results

There are many question answer sites are available now a day. In that Community Question Answering Sites are achieved lots of popularity over last years. But the drawback of available question answering system is that it can only provide multiple solutions of the textual answer and user need to select one of them. In this paper, we propose a scheme that enriches the textual answer with multimedia data. Our scheme consists of four models: QA pair extraction, answer medium selection, query generation and selection and presentation. The type media information added with the textual data is determined. The question answer pair is generated from the available Community Question Answering Sites database. Query is generated for the multimedia data. The final resulting data must be selected after re-ranking and removal operation and then present to the user. Keyword-Qa pair, multimedia selection, query generation, openCV, surf

Knowledge fixation and accretion: longitudinal analysis of a social question-answering site

Purpose– The purpose of this paper is to investigate longitudinal features of an established social question-answering (Q&A) site to study how question-answer resources and other community features change over time.Design/methodology/approach– Statistical analysis and visualisation was performed on the full data dump from the Stack Overflow social Q&A site for programmers.Findings– The timing of answers is as strong a predictor of acceptance – a proxy for user satisfaction – as the structural features of provided answers sometimes associated with quality. While many questions and answer exchanges are short-lived, there is a small yet interesting subset of questions where new answers receive community approval and which may end up being ranked more highly than early answers.Research limitations/implications– As a large-scale data oriented research study, this work says little about user motivations to find and contribute new knowledge to old questions or about the impact of the resource on the consumer. This will require complementary studies using qualitative and evaluative methods.Practical implications– While content contribution to social question-asking is largely undertaken within a very short time frame, content consumption is usually over far longer periods. Methods and incentives by which content can be updated and maintained need to be considered. This work should be of interest to knowledge exchange community designers and managers.Originality/value– Few studies have looked at temporal patterns in social Q&A and how time and the moderation and voting systems employed may shape resource quality.

A Supervised Keyphrase Extraction Method Based on the Logistic Regression Model for Social Question Answering Sites

This paper proposes a supervised machine learning method for the problem of automatic keyphrase extraction for Social Question Answering (SQA) sites. The method is developed by: 1) Analyzing the structural and activity characteristics of typical SQA sites, 2) Developing and categorizing four types of calculation features that can describe those characteristics, and 3) Developing customized logistic regression model to be trained by the real dataset from six popular SQA sites, in both English and Chinese. Experimental results show the influences from those proposed SQA related features vary, some are helpful to keyphrase extraction for SQA sites of both languages while some are only useful for a specific site. The results also demonstrate a generally better performance comparing to a typical keyphrase extraction algorithms published previously like KEA.

Incentivizing High-Quality Content from Heterogeneous Users: On the Existence of Nash Equilibrium

We study the existence of pure Nash equilibrium (PNE) for the mechanisms used in Internet services (e.g., online reviews and question-answering websites) to incentivize users to generate high-quality content. Most existing work assumes that users are homogeneous and have the same ability. However, real-world users are heterogeneous and their abilities can be very different from each other due to their diversity in background, culture, and profession. In this work, we consider the following setting: (1) the users are heterogeneous and each of them has a private type indicating the best quality of the content he/she can generate; (2) all the users share a fixed total reward. With this setting, we study the existence of pure Nash equilibrium of several mechanisms composed by different allocation rules, action spaces, and information availability. We prove the existence of PNE for some mechanisms and the non-existence for some other mechanisms. We also discuss how to find a PNE (if exists) through either a constructive way or a search algorithm.

Great Question! Question Quality in Community Q&A

Asking the right question in the right way is an art (and a science). In a community question-answering setting, a good question is not just one that is found to be useful by other people: a question is good if it is also presented clearly and shows prior research. Using a community question-answering site that allows voting over the questions, we show that there is a notion of question quality that goes beyond mere popularity. We present techniques using latent topic models to automatically predict the quality of questions based on their content. Our best system achieves a prediction accuracy of 72%, beating out strong baselines by a significant amount. We also examine the effect of question quality on the dynamics of user behavior and the longevity of questions.

Ranking potential reply-providers in community question answering system

Community Question Answering (CQA) websites have greatly facilitated users' lives, with an increasing number of people seeking help and exchanging ideas on the Internet. This newly-emerged community features two characteristics: social relations and an ask-reply mechanism. As users' behaviours and social statuses play a more important role in CQA services than traditional answer retrieving websites, researchers' concerns have shifted from the need to passively find existing answers to actively seeking potential reply providers that may give answers in the near future. We analyse datasets derived from an online CQA system named “Quora”, and observed that compared with traditional question answering services, users tend to contribute replies rather than questions for help in the CQA system. Inspired by the findings, we seek ways to evaluate the users' ability to offer prompt and reliable help, taking into account activity, authority and social reputation characteristics. We propose a hybrid method that is based on a Question-User network and social network using optimised PageRank algorithm. Experimental results show the efficiency of the proposed method for ranking potential answer-providers.

An Effective Method for Habitual Behavior Extraction from the Internet

Many research studies are being conducted about the analysis of human behavior using sensor devices in the real world, and a variety of information can be found all over Internet. The primary objective is to improve social behavior and habits, such as the prohibition against smoking and the use mobile phones while driving. These unhealthy social behaviors and habits tend to cause health problems and antisocial behaviors. Behavioral modification specialists understand that habitual behavior is one of the most important behaviors in solving these issues. This paper proposes a new method to extract habitual behaviors for discovering the objectives of the behavioral modification. Specifically, Latent Dirichlet Allocation, or LDA, is used for clustering words into appropriate topics of periodical behaviors from literary expressions, and Point- wise Mutual Information, or PMI, is applied to select suitable words for habitual behaviors. The technique by using text data from question-answering websites from the telecommunications industry area was evaluated and showed good performance results.

Integrating expert profile, reputation and link analysis for expert finding in question-answering websites

Question answering websites are becoming an ever more popular knowledge sharing platform. On such websites, people may ask any type of question and then wait for someone else to answer the question. However, in this manner, askers may not obtain correct answers from appropriate experts. Recently, various approaches have been proposed to automatically find experts in question answering websites. In this paper, we propose a novel hybrid approach to effectively find experts for the category of the target question in question answering websites. Our approach considers user subject relevance, user reputation and authority of a category in finding experts. A user’s subject relevance denotes the relevance of a user’s domain knowledge to the target question. A user’s reputation is derived from the user’s historical question-answering records, while user authority is derived from link analysis. Moreover, our proposed approach has been extended to develop a question dependent approach that considers the relevance of historical questions to the target question in deriving user domain knowledge, reputation and authority. We used a dataset obtained from Yahoo! Answer Taiwan to evaluate our approach. Our experiment results show that our proposed methods outperform other conventional methods.

Finding similar questions in collaborative question answering archives: toward bootstrapping-based equivalent pattern learning

Many questions submitted to Collaborative Question Answering (CQA) sites have similar questions answered before. We propose a precise approach of automatically finding an answer to such questions by automatically identifying "equivalent" questions submitted and answered, in the past. Our method is based on automatically generating equivalent question patterns by grouping together questions that have previously obtained the same answers. The generated patterns are used as seed patterns to match more questions to extract large number of equivalent patterns by a new bootstrapping-based learning method. The resulting patterns can be applied to match a new question to an equivalent one that has already been answered, and thus suggest potential answers automatically. We experimented with this approach over a large collection of more than 200,000 real questions drawn from the Yahoo! Answers archive, automatically acquiring over 16,991 groups of equivalent question patterns. These patterns allow our method to obtain over 57% recall and over 54% precision on suggesting an answer automatically to new questions, significantly improving over baseline methods.

A comparative assessment of answer quality on four question answering sites

Question answering (Q&A) sites, where communities of volunteers answer questions, may provide faster, cheaper, and better services than traditional institutions. However, like other Web 2.0 platforms, user-created content raises concerns about information quality. At the same time, Q&A sites may provide answers of different quality because they have different communities and technological platforms. This paper compares answer quality on four Q&A sites: Askville, WikiAnswers, Wikipedia Reference Desk, and Yahoo! Answers. Findings indicate that: (1) similar collaborative processes on these sites result in a wide range of outcomes, and significant differences in answer accuracy, completeness, and verifiability were evident; (2) answer multiplication does not always result in better information; it yields more complete and verifiable answers but does not result in higher accuracy levels; and (3) a Q&A site’s popularity does not correlate with its answer quality, on all three measures.

Community QA Question Classification: Is the Asker Looking for Subjective Answers or Not?

Community question answering (CQA) sites such as Yahoo! Chiebukuro are known to be very useful resources for automatic question answering (QA) systems. However, CQA users often post questions expecting not some general truths but rather opinions of different people. We believe that a QA system should act according to these different question types. We therefore define two question types based on whether the questioner expects subjective or objective answers, and report on an automatic question classification experiment. We achieve over 80% weighted accuracy using uni-gram and bi-gram features learned by Naive Bayes with smoothing. We also discuss the inter-annotator agreement and its impact on automatic classification accuracy, as well as what kind of questions tend to be misclassified.