What We Vote for? Answer Selection from User Expertise View in Community Question Answering

Lyu, Shanshan; Ouyang, Wentao; Wang, Yongqing; Shen, Huawei; Cheng, Xueqi

doi:10.1145/3308558.3313510

Cited by 16 publications

(18 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Recent research topics regarding cQA users relate to modeling their areas of expertise and quantifying their impacts on answer selection [3]; how the evolution of their roles in the community impacts the content relevance between the answerer and the question [4]; the intimacy between askers and answerers [5]. From another angle, current research has focused its attention on discovering informative features of genuine experts [6][7][8], and discovering patterns of interactions between community fellows [9]. Contrary to the vast bulk of recent research, we take the lead on studying the challenges faced by supervised models when discovering discriminative patterns of the age demographics of cQA members.…”

Section: Related Workmentioning

confidence: 99%

Coming to Grips with Age Prediction on Imbalanced Multimodal Community Question Answering Data

2021

View full text Add to dashboard Cite

For almost every online service, it is fundamental to understand patterns, differences and trends revealed by age demographic analysis—for example, take the discovery of malicious activity, including identity theft, violation of community guidelines and fake profiles. In the particular case of platforms such as Facebook, Twitter and Yahoo! Answers, user demographics have impacts on their revenues and user experience; demographics assist in ensuring that the needs of each cohort are fulfilled via personalizing and contextualizing content. Despite the fact that technology has been made more accessible, thereby becoming evermore prevalent in both personal and professional lives alike, older people continue to trail Gen Z and Millennials in its adoption. This trailing brings about an under-representation that has a harmful influence on the demographic analysis and on supervised machine learning models. To that end, this paper pioneers attempts at examining this and other major challenges facing three distinct modalities when dealing with community question answering (cQA) platforms (i.e., texts, images and metadata). As for textual inputs, we propose an age-batched greedy curriculum learning (AGCL) approach to lessen the effects of their inherent class imbalances. When built on top of FastText shallow neural networks, AGCL achieved an increase of ca. 4% in macro-F1-score with respect to baseline systems (i.e., off-the-shelf deep neural networks). With regard to metadata, our experiments show that random forest classifiers significantly improve their performance when individuals close to generational borders are excluded (up to 20% more accuracy); and by experimenting with neural network-based visual classifiers, we discovered that images are the most challenging modality for age prediction. In fact, it is hard for a visual inspection to connect profile pictures with age cohorts, and there are considerable differences in their group distributions with respect to meta-data and textual inputs. All in all, we envisage that our findings will be highly relevant as guidelines for constructing assorted multimodal supervised models for automatic age recognition across cQA platforms.

show abstract

Section: Related Workmentioning

confidence: 99%

Coming to Grips with Age Prediction on Imbalanced Multimodal Community Question Answering Data

2021

View full text Add to dashboard Cite

show abstract

“…Highest upvoted comments represent community consensus on the most trustworthy response for the post [16]. In particular, we rank comments for each post m, in the order of descending cosine similarity between their embedding, a m,n , and the latent trustworthy comment embeddings, a * m .…”

Section: Trustworthy Comment Embedding Analysismentioning

confidence: 99%

CrowdQM: Learning Aspect-Level User Reliability and Comment Trustworthiness in Discussion Forums

Morales

Narang

Sundaram

et al. 2020

Advances in Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Community discussion forums are increasingly used to seek advice; however, they often contain conflicting and unreliable information. Truth discovery models estimate source reliability and infer information trustworthiness simultaneously in a mutual reinforcement manner, and can be used to distinguish trustworthy comments with no supervision. However, they do not capture the diversity of word expressions and learn a single reliability score for the user. CrowdQM addresses these limitations by modeling the fine-grained aspect-level reliability of users and incorporate semantic similarity between words to learn a latent trustworthy comment embedding. We apply our latent trustworthy comment for comment ranking for three diverse communities in Reddit and show consistent improvement over non-aspect based approaches. We also show qualitative results on learned reliability scores and word embeddings by our model.A. Morales and K. Narang-Equal Contribution.

show abstract

“…It is loved and sought after by the majority of users. Typical CQA websites include StackOverflow, Yahoo Answers, TurboTax [1][2][3][4][5][6]. At present, industry peers and researchers in related fields have gradually begun to pay attention to questions in the CQA, and continue to enrich the research in the field of the CQA.…”

Section: Introductionmentioning

confidence: 99%

Answer Classification via Machine Learning in Community Question Answering

Jiang¹,

Zhang²,

Jia³

et al. 2021

Journal on Artificial Intelligence

View full text Add to dashboard Cite

As a new type of knowledge sharing platform, the community question answer website realizes the acquisition and sharing of knowledge, and is loved and sought after by the majority of users. But for multi-answer questions, answer quality assessment becomes a challenge. The answer selection in CQA (Community Question Answer) was proposed as a challenge task in the SemEval competition, which gave a data set and proposed two subtasks. Task-A is to give a question (including short title and extended description) and its answers, and divide each answer into absolutely relevant (good), potentially relevant (potential) and bad or irrelevant (bad, dialog, non-English, other). Task-B is to give a YES/NO type question (including short title and extended description) and some answers. Based on the answer of the absolute correlation type (good), judge whether the answer to the whole question should be yes, no or uncertain. This paper first preprocesses this data set, and then uses natural language processing technology to perform word segmentation, part-of-speech tagging and named entity recognition on the data set, and then perform feature extraction on the preprocessed data set. Finally, SVM and random forest are used to classify on the basis of feature extraction, and the classification results are analyzed and compared. The experiments in this paper show that SVM and random forest methods have good results on the data set, and exceed the multi-classifier ensemble learning method and hierarchical classification method proposed by the predecessors.

show abstract

What We Vote for? Answer Selection from User Expertise View in Community Question Answering

Cited by 16 publications

References 31 publications

Coming to Grips with Age Prediction on Imbalanced Multimodal Community Question Answering Data

Coming to Grips with Age Prediction on Imbalanced Multimodal Community Question Answering Data

CrowdQM: Learning Aspect-Level User Reliability and Comment Trustworthiness in Discussion Forums

Answer Classification via Machine Learning in Community Question Answering

Contact Info

Product

Resources

About