Joint voting prediction for questions and answers in CQA

Yao, Yuan; Tong, Hanghang; Xie, Tao; Akoglu, Leman; Xu, Feng; Lü, Jian

doi:10.1109/asonam.2014.6921607

Cited by 8 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These works used textual and nontextual features to evaluate quality. Some research also evaluated answer quality based on question quality, finding a high correlation between the two (Yao et al, ; Arora, Ganguly, & Jones, ). Different advance machine learning techniques such as random forest and deep neural network are applied to find the quality of contents in CQA and the users who fail to generate the good content (Chen et al, ; Le, Shah, & Choi, ).…”

Section: Background and Related Workmentioning

confidence: 99%

Retrieving people: Identifying potential answerers in Community Question‐Answering

Shah

2018

Asso for Info Science & Tech

View full text Add to dashboard Cite

Community Question-Answering (CQA) sites have become popular venues where people can ask questions, seek information, or share knowledge with a user community. Although responses on CQA sites are obviously slower than information retrieved by a search engine, one of the most frustrating aspects of CQAs occurs when an asker's posted question does not receive a reasonable answer or remains unanswered. CQA sites could improve users' experience by identifying potential answerers and routing appropriate questions to them. In this paper, we predict the potential answerers based on question content and user profiles. Our approach builds user profiles based on past activity. When a new question is posted, the proposed method computes scores between the question and all user profiles to find the potential answerers. We conduct extensive experimental evaluations on two popular CQA sites -Yahoo! Answers and Stack Overflow -to show the effectiveness of our algorithm. The results show that our technique is able to predict a small group of 1000 users from which at least one user will answer the question with a probability higher than 50% in both CQA sites. Further analysis indicates that topic interest and activity level can improve the correctness of our approach.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

Retrieving people: Identifying potential answerers in Community Question‐Answering

Shah

2018

Asso for Info Science & Tech

View full text Add to dashboard Cite

show abstract

“…Very few studies have quantitatively examined the part-whole relationships for the purpose of prediction. In CQA sites, empirical studies have shown a strong correlation between the question voting score and the average/maximum answer voting score [22]. Based on this observation, a joint predictive model that leverages question/answer coupling is proposed that is also able to capture the dynamics of the community posts [23].…”

Section: Related Workmentioning

confidence: 99%

“…First (Modeling Challenge), the relationship between the parts outcome and whole outcome might be complicated, beyond the simple addition or linear combination. For example, the authors in [22] empirically identi ed a non-linear correlation between the impacts of questions and the associated answers, that is, the impact of a question is much more strongly correlated with that of the best answer it receives, compared with the average impact of its associated answers. However, how to leverage such non-linear relationship between the parts and whole outcome has largely remained open.…”

Section: Introductionmentioning

confidence: 99%

Is the Whole Greater Than the Sum of Its Parts?

Tong

Wang

et al. 2017

Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Self Cite

View full text Add to dashboard Cite

e relationship routinely nds itself in many disciplines, ranging from collaborative teams, crowdsourcing, autonomous systems to networked systems. From the algorithmic perspective, the existing work has primarily focused on predicting the outcomes of the whole and parts, by either separate models or linear joint models, which assume the outcome of the parts has a linear and independent e ect on the outcome of the whole. In this paper, we propose a joint predictive method named PAROLE to simultaneously and mutually predict the part and whole outcomes. e proposed method o ers two distinct advantages over the existing work. First (Model Generality), we formulate joint outcome prediction as a generic optimization problem, which is able to encode a variety of complex relationships between the outcome of the whole and parts, beyond the linear independence assumption. Second (Algorithm E cacy), we propose an e ective and e cient block coordinate descent algorithm, which is able to nd the coordinate-wise optimum with a linear complexity in both time and space. Extensive empirical evaluations on real-world datasets demonstrate that the proposed PAROLE (1) leads to consistent prediction performance improvement by modeling the non-linear part-whole relationship as well as part-part interdependency, and (2) scales linearly in terms of the size of the training dataset.

show abstract

“…In this article, we focus on the voting score prediction of questions/answers shortly after they are posted in the CQA sites. Such a task is essential for the prosperity and sustainability of the CQA ecosystem, and it may benefit all types of users, including the information producers and consumers [2]. For example, detecting potentially high-score answers can benefit the questioners as well as the people who have similar questions; it would also be helpful to identify high-score questions in the early stage and route them to expert answerers.…”

Section: Introductionmentioning

confidence: 99%

“…For example, detecting potentially high-score answers can benefit the questioners as well as the people who have similar questions; it would also be helpful to identify high-score questions in the early stage and route them to expert answerers. Generally speaking, there are three key aspects that matter with the voting prediction of a post, namely, (1) the non-linearity between features and output, (2) the coupling between questions and answers, and (3) the dynamics (of training data sets). First, both the contextual features (e.g., the reputation of the user who issues the question, etc.)…”

Section: Introductionmentioning

confidence: 99%

Routing Queries to the Right Expert in Online CQA

Bharathi¹

2019

IJRASET

View full text Add to dashboard Cite

The popular websites such as Stack Overflow and Yahoo provides the question and answer which is posted by the anonymous user by all over the world. In these sites, the programmers from all over the world will answer for the queries which are posted on the websites. In the existing system, if any user wants to know the exact answer they have to mine the answer between the thousands of answer posted on the website and still it was very complicated to know the exact answer and it was very time consuming. So that user has to apply all the answer to get the exact answer. So to overcome from this situation, we proposing the technique in which user can view the exact answer from the websites as it will be displayed as a first answer for that question through post voting prediction. In the voting process, the voting will be given based on the conditions. If that answer satisfies the conditions, it will be up voted. In this way, our proposed technique provides the effectiveness and efficiency, optimality, correctness and complexity.

show abstract

Joint voting prediction for questions and answers in CQA

Cited by 8 publications

References 15 publications

Retrieving people: Identifying potential answerers in Community Question‐Answering

Retrieving people: Identifying potential answerers in Community Question‐Answering

Is the Whole Greater Than the Sum of Its Parts?

Routing Queries to the Right Expert in Online CQA

Contact Info

Product

Resources

About