Community Question Research Articles

Cross-language question retrieval with multi-layer representation and layer-wise adversary

In cross-language question retrieval (CLQR), users employ a new question in one language to search the community question answering (CQA) archives for similar questions in another language. In addition to the ranking problem in monolingual question retrieval, one needs to bridge the language gap in CLQR. The existing adversarial models for cross-language learning normally rely on a single adversarial component. Since natural languages consist of units of different abstract levels, we argue that crossing the language gap adaptatively on different levels with multiple adversarial components should lead to smoother text representation and better CLQR performance. To this end, we first encode questions into multi-layer representations of different abstract levels with a CNN based model which enhances conventional models with diverse kernel shapes and the corresponding pooling strategy so as to capture different aspects of a text segment. We then impose a set of adversarial components on different layers of question representation so as to decide the appropriate abstract levels and their role in performing cross-language mapping. Experimental results on two real-world datasets demonstrate that our model outperforms state-of-the-art models for CLQR, which is on par with the strong machine translation baselines and most monolingual baselines.

Truth without Reconciliation: A Human Rights History of Ghana, Abena Ampofoa Asare

Reviewed by: Truth without Reconciliation: A Human Rights History of Ghana by Abena Ampofoa Asare Jeffrey Ahlman BOOK REVIEW of Asare, Abena Ampofoa. 2018. Truth without Reconciliation: A Human Rights History of Ghana. Philadelphia: University of Pennsylvania Press. 243 pp. $79.95 (cloth). Abena Ampofoa Asare’s Truth without Reconciliation is a difficult book; it is also an extremely powerful book. Framed around the National Reconciliation Commission (NRC) established in 2002 by the New Patriotic Party government of President J. A. Kufuor, it forces a rereading of Ghana’s postindependence history from the perspective of Ghanaian men and women harmed by the country’s postcolonial governments. At its heart is the methodological question of how scholars are to make sense of the archive created by the NRC. Eschewing a straightforward reading of the archive, Asare skillfully guides her readers through many nuances, hidden meanings, and contradictions that mark the petitions and testimonies collected by the NRC. In doing so, she untangles questions of power, gender, memory, family, and community, creating a compendium of narratives of Ghanaian postcolonial history. The narratives that she constructs, however, do not necessarily turn on their head the conventional narratives that surround the country’s recent history, but instead personalize and complicate them, injecting into them the depth and messiness of personal experience often lost in scholarly discourse. Asare opens the book with a short history surrounding the creation of the NRC and the partisan controversies it generated. She moves on to chapters focused on the violence inflicted on market women, families’ experiences with detention and political violence, the nonjurisdictional claims of those who lost property and their livelihoods to the country’s development schemes, and the blurred lines between perpetrators and victims in the stories of soldiers and citizens. Each of these chapters is organized around the personal accounts of individuals who petitioned and came before the NRC. Asare is not only meticulous in her presentation of these individuals’ stories, but deeply empathetic, as she weaves the individuality of each petitioner’s narrative into the complicated scaffolding of Ghana’s national histories. For instance, in her chapter on market women (chapter three), she turns to the long history of demonization the country’s market women have faced at the hands of its postcolonial governments, for which the market women regularly served as scapegoats for national economic troubles, leading most tragically to the 1979 burning of Accra’s Makola Market and additional attacks on markets in other major Ghanaian cities and towns (75). What Asare adds to this narrative is not simply an accounting of the [End Page 137] injuries sustained by these women due to the government’s actions: more importantly, she emphasizes the ways in which the market women who petitioned the NRC understood and articulated the social and professional bonds shaping their roles as traders, women, mothers, and caregivers—roles disrupted and, in some cases, destroyed by governmental actions. Exhibited in all of Truth without Reconciliation’s chapters is a rethinking of what a history of human rights could or should look like. For Asare, the question of human rights history is ultimately a methodological question: it is a history that extends beyond the institutional artifice of truth-and-reconciliation commissions like the NRC to the act of archival creation, negotiation, access, and silence, for embedded within archives like that of the NRC are the voices of those affected by governments’ actions. As Asare emphasizes throughout her book, too often questions of human rights are constrained to vague and abstract platitudes, yet, as she shows, a much richer and more authentic contestation of what human rights comprise is found in the archival voices of those who have come before commissions like the NRC, as individuals from nearly all walks of life engaged in debates over the rights and obligations shaping their relationships with the state. Asare’s book is thus a must-read for any student of Ghanaian history, as well as those interested in the history of human rights. It is one of those rare books whose theoretical and methodological interventions are equally matched by their empirical rigor. Jeffrey Ahlman Smith College Copyright © 2019 The Trustees of Indiana University

Interrogating the relationship between language, thought versus individual and community: a conversation with Agada and Egbai

In this short essay, I will converse with Ada Agada and Uti Egbai on their article titled “Language, Thought, and Interpersonal Communication: A Cross-Cultural Conversation on the Question of Individuality and Community,” published in [Filosofia Theoretica: Journal of African Philosophy, Religions and Culture Vol 7, No 2, 141 -161, 2018]. I will articulate the major contributions of the authors and critically engage their ideas in order to open new vistas for thought. I contend that the relationship that exists between language and thought is like the relationship that exists between individual and community in the debate of personhood in African philosophy. This relationship is what I call “arumaristic relationship”.Keywords: Personhood, Individual, Community, Language, Thought, Agada, Egbai

Contextual Documentation Referencing on Stack Overflow

Software engineering is knowledge-intensive and requires software developers to continually search for knowledge, often on community question answering platforms such as Stack Overflow. Such information sharing platforms do not exist in isolation, and part of the evidence that they exist in a broader software documentation ecosystem is the common presence of hyperlinks to other documentation resources found in forum posts. With the goal of helping to improve the information diffusion between Stack Overflow and other documentation resources, we conducted a study to answer the question of how and why documentation is referenced in Stack Overflow threads. We sampled and classified 759 links from two different domains, regular expressions and Android development, to qualitatively and quantitatively analyze the links’ context and purpose, including attribution, awareness, and recommendations. We found that links on Stack Overflow serve a wide range of distinct purposes, ranging from citation links attributing content copied into Stack Overflow, over links clarifying concepts using Wikipedia pages, to recommendations of software components and resources for background reading. This purpose spectrum has major corollaries, including our observation that links to documentation resources are a reflection of the information needs typical to a technology domain. We contribute a framework and method to analyze the context and purpose of Stack Overflow links, a public dataset of annotated links, and a description of five major observations about linking practices on Stack Overflow. Those observations include the above-mentioned purpose spectrum, its interplay with documentation resources and applications domains, and the fact that links on Stack Overflow often lack context in form of accompanying quotes or summaries. We further point to potential tool support to enhance the information diffusion between Stack Overflow and other documentation resources.

A Comprehensive Retrospection of Literature Reported Works of Community Question Answering Systems

Community question answering CQA) systems are rapidly gaining attention in the society. Several researchers have actively engaged in improving the theories associated with question answering (QA) systems. This paper reviews the literature reported works on question answering QA systems. In this paper, we discuss on the early contributions on QA systems along with their present and future scope. We have categorized the literature reported works into 20 subgroups according to their significance and relevance. The works in each group will be brought out along with their inter-relevance. Finding the question and answer quality is the prime challenge almost addressed by many researchers. Modeling similar questions, identifying experts in prior and understanding seeker satisfaction also considered as potential challenges. Researchers at the most have done experimentations on popular CQAs like Yahoo! Answers, Wiki Answers, Baidu Knows, Brianly, Quora, Pubmed and Stack Overflow respectively. Machine learning, probabilistic modeling, deep learning and hybrid approach of solving show profound significance in addressing various challenges encounter with QA systems. Today the paradigm of CQA systems took the shift by serving as Open Educational Resources to learning community.

Extracting core answers using the grey wolf optimizer in community question answering

A rapidly increasing number of users in community question answering has led to an explosive growth of answers. Thus, it is becoming increasingly more difficult to browse all the answers. Choosing a subset of answers arbitrarily will likely lead to cognitive bias and even poor decisions. Reading a set of core answers that can cover most topics of all the answers is a novel method to overcome information overload and facilitates information retrieval. In this paper, the method named AnsExt for extracting core answers using a kind of swarm intelligence algorithm, named the grey wolf optimizer (GWO), is proposed. First, answers are modeled with the biterm topic model, which fits both short and long texts. Then, factors including quality, coverage, redundancy and number in extracting core answers are defined. Two scenarios are modeled with the requirements of quality, coverage and redundancy: extracting the least number of answers and extracting a predefined number of answers. To extract the minimum number of core questions, a binary GWO is used to resolve the single-objective optimal model. The binary multi-objective GWO is constructed to resolve the optimal model, which is used to extract a predefined number of core answers. Extensive experiments are conducted on real datasets. The results show that the proposed method is feasible and performs well.

The corporate nationality: A question of culture and community?

Many scholars in management and business history are using nationality as a core criterion to classify corporations and to explain their behaviour, values, and strategy. In this perspective, being born in a specific country defined by its frontiers and institutions explicates the values and behaviour of its inhabitants—and companies within this context are described as more ‘German’, ‘Italian’, or ‘French’. However, it does not help to analyse in depth the complex and various differences between what is observable in day-to-day business life and what is in underneath. A better understanding of corporation nationality imposes to use the concept of culture. It implies to underline the fact that ‘nation’ is often delimitated by frontiers and most of the time is the result of political choices. If nationality explains some element of corporate culture, it cannot explain all cultural dimensions. One reason is that culture does not stop at national frontiers. This article has several aims: (1) it discusses the use of nationality as a means to describe and analyse the organizational behaviour of companies and their communities; (2) it interrogates ‘nationality’ as a powerful factor which enables to aggregate different communities within large corporations; and (3) it analyzes how ‘nationality’ is influencing managerial practices. Most of the time, ‘nationality’ is mobilized to describe the global strategy or practices of big corporations. The author argues that nationality is a crucial factor for corporate culture in many ways—not as a ‘given’ fact but as a larger part of the institutional design of a company (and its development).

Building interactive sentence-aware representation based on generative language model for community question answering

Semantic matching between question and answer sentences involves recognizing whether a candidate answer is relevant to a particular input question. Given the fact that such matching does not examine a question or an answer individually, context information outside the sentence should be considered equally important to the within-sentence syntactic context. This motivates us to design a new question-answer matching model, built upon a cross-sentence, context-aware, bi-directional long short-term memory architecture. The interactive attention mechanisms are proposed which automatically select salient positional sentence representations, that contribute more significantly towards the relevance between two question and answer. A new quantity called context information jump is proposed to facilitate the formulation of the attention weights, and is computed via the joint states of adjacent words. An interactive-aware sentence representation is constructed by connecting a combination of multiple sentence positional representations to each hidden representation state. In the experiments, the proposed method is compared with existed models, using four public community datasets, and the evaluations show that it is very competitive. In particular, it offers 0.32%-1.8% improvement over the best performing model for three out of four datasets, while for the remaining one performance is around 0.2% of the best performer.

The topic of terrorism on Yahoo! Answers: questions, answers and users’ anonymity

PurposeThe purpose of this paper is to explore the use of community question answering sites (CQAs) on the topic of terrorism. Three research questions are investigated: what are the dominant themes reflected in terrorism-related questions? How do answer characteristics vary with question themes? How does users’ anonymity relate to question themes and answer characteristics?Design/methodology/approachData include 300 questions that attracted 2,194 answers on the community question answering Yahoo! Answers. Content analysis was employed.FindingsThe questions reflected the community’s information needs ranging from the life of extremists to counter-terrorism policies. Answers were laden with negative emotions reflecting hate speech and Islamophobia, making claims that were rarely verifiable. Users who posted sensitive content generally remained anonymous.Practical implicationsThis paper raises awareness of how CQAs are used to exchange information about sensitive topics such as terrorism. It calls for governments and law enforcement agencies to collaborate with major social media companies to develop a process for cross-platform blacklisting of users and content, as well as identifying those who are vulnerable.Originality/valueTheoretically, it contributes to the academic discourse on terrorism in CQAs by exploring the type of questions asked, and the sort of answers they attract. Methodologically, the paper serves to enrich the literature around terrorism and social media that has hitherto mostly drawn data from Facebook and Twitter.

“A WeeklyMäshräpto Tackle Extremism”: Music-Making in Uyghur Communities and Intangible Cultural Heritage in China

AbstractThis article provides an in-depth study and critique of the nomination and inscription of an item on UNESCO’s lists of intangible cultural heritage and the developments following its acceptance. China is now a major partner in UNESCO’s heritage projects, but the application and experience of heritage initiatives across China have been highly uneven. I discuss the particular challenges presented in the contested, predominantly Muslim Xinjiang Uyghur Autonomous Region. I focus on the question of community, a term that lies at the heart of the UNESCO literature. What do we mean by community, and why do we think it matters? How does socially embedded music-making facilitate community, and how do heritage initiatives change that?مهزكۇر ماقاله تهنقﯩدﯨي مهۋقهده تۇرۇپ، بﯩر غهيرى رهسمﯩي مهدهنﯩيهت مﯩراسﯩنﯩڭ بﯩرلهشكهن دۆلهتلهر تهشكﯩلاتى پهن مهدهنﯩيهت مهھكﯩمﯩسﯩگه قانداق قﯩلﯩپ نامزاتلﯩققا كۆرسﯩتﯩلگهنلﯩكى ۋه ئهنگه ئېلﯩنغانلﯩقى، ھهم بۇ مهدهنﯩيهت مﯩراسى قوبۇل قﯩلﯩنغاندﯨن كېيﯩنكى تهرهققﯩيات ئهھۋاللﯩرى قاتارلﯩق مهسﯩلﯩلهر ئۈستﯩده ئېلﯩپ بېرﯨلغان ئﯩنچﯩكه تهتقﯩقات مهزمۇنلﯩرى بﯩلهن تهمﯩنلهيدۇ. گهرچه خﯩتاي ھۆكۈمﯩتى ھازﯨر بﯩرلهشكهن دۆلهتلهر تهشكﯩلاتى پهن مهدهنﯩيهت مهھكﯩمﯩسﯩنﯩڭ مهدهنﯩي مﯩراسلارنى قوغداش ئﯩش تۈرلﯩرﯨگه ئهڭ كوپ ئﯩقتﯩسادﯨي ياردهم بېرﯨدﯨغان دۆلهت بولسﯩمۇ ئهمما خﯩتاي دۆلﯩتﯩده مهدهنﯩيهت مﯩراسلﯩرﯨنى قوغداشقا ئﯩلتﯩماس قﯩلﯩش ۋه ئۇنﯩڭدﯨن كېيﯩنكى خﯩزمهتلهرنﯩڭ ئﯩجرا قﯩلﯩنﯩشى ئﯩنتايﯩن تهڭپۇڭسﯩز. مهن مۇشۇ تېمﯩنى چۆرﯨدﯨگهن ھالدا، بولۇپمۇ مۇسۇلمانلار مۇتلهق كۆپ سانلﯩقنى ئﯩگهللهيدﯨغان شﯩنجاڭ ئۇيغۇر ئاپتونوم رايونﯩدﯨكى مهۋجۇت مهسﯩلﯩلهر ھهققﯩده مۇنازﯨره ئېلﯩپ بارﯨمهن. مهن، بۇ يهردﯨكى جهمﯩيهتته بولۇۋاتقان مهسﯩلﯩلهرنى ھالقﯩلﯩق نۇقتا قﯩلﯩپ تۇرۇپ، بﯩرلهشكهن دۆلهتلهر تهشكﯩلاتى پهن مهدهنﯩيهت مهھكﯩمﯩسﯩنﯩڭ ئﯩچﯩده قانچﯩلﯩك ساختﯩلﯩقلارنﯩڭ بولۇۋاتقانلﯩقﯩنى ئوتتۇرﯨغا قويﯩمهن. بۇ ئهھۋاللار ئۇيغۇر جهمئﯩيﯩتى تهرﯨپﯩدﯨن قانداق چۈشﯩنﯩلﯩدۇ ۋه بﯩز نﯩمﯩشقا بۇ ئﯩشنى ناھايﯩتى ئېغﯩر مهسﯩله دهپ قارايمﯩز؟ مهدهنﯩيهت مﯩراسلﯩرﯨغا ۋارﯨسلﯩق قﯩلﯩش خﯩزمهتلﯩرى شۇ جهمئﯩيهتنﯩڭ ئﯩجتﯩمائﯩي ساھهسى بﯩلهن زﯨچ باغلﯩنﯩشلﯩق بولغان مۇزﯨكا ئﯩشلهپچﯩقﯩرﯨشﯩغا قانداق ئۆزگﯩرﯨشلهرنى ئېلﯩپ كېلﯩدۇ؟

India in The Pakistan Times, September 1950-February 1951

ABSTRACTThis article offers a glimpse of the representation of India in The Pakistan Times in 1950–51, in the last months of the first phase of its editorship by Faiz Ahmed Faiz. It contextualizes this reflection in a period of relative lull, marked by pacts on minorities and trade, within the post-Partition upheaval. Focusing especially on editorials, commentaries, and cartoons, it highlights the formation of views of the neighbor from the news that was being generated. The stress is on firstly, a variety of themes included – not just the “unfinished business of Partition”, secondly, a responsible review of these themes – not just a narrow-minded nationalist reading, and thirdly, an involved – if not intertwined – interpretation of both countries’ socio-economic relations, while not underplaying the differences of their regimes. Further, the article goes beyond an inter-governmental framework, as The Pakistan Times went beyond it and presents a slice of the possibilities then existing, of a progressive politics on questions of class, community, and capital. In early independent Pakistan, this was personified by Faiz and this article, while not on him nor on an extensive Saidian reflection of the “other”, tries to trace his imprint in the representation of India in the newspaper he edited and seeks to contribute to our understanding of India as constructed through Pakistani eyes.

Gender and Religious life in French Revolutionary Drama by Annelle Curulla

Reviewed by: Gender and Religious life in French Revolutionary Drama by Annelle Curulla Cecilia Feilla GENDER AND RELIGIOUS LIFE IN FRENCH REVOLUTIONARY DRAMA. By Annelle Curulla. Oxford University Studies in the Enlightenment series. Oxford: Oxford University Press, 2018; pp. 216. Among the sweeping theatre reforms of the French Revolution was the historic lifting of the age-old ban on the depiction of religious characters, attire, rituals, and settings onstage. The effect on the repertory was immediate, as a host of new productions brought the world of the cloister onto the French stage for the first time (ironically, just as the revolutionary government was passing legislation to ban monastic vows and close convents and monasteries). In her meticulously researched and clearly written study of this phenomenon, Annelle Curulla finds in the stage representation of the religious a trope for thinking through the larger changes occurring during the French Revolution. She argues that religious representations were not mere novelties that fascinated audiences, but rather productive sites for dialogue about secularization and for innovation in costume, setting, and subject matter. Curulla narrows her scope to monastic plays (or pièces monacales in French), which she reads alongside multiple contexts—political, social, cultural, literary, gendered, and theatrical—in order to elucidate how stories of monks and nuns allowed playwrights to address broader questions of identity and community as they were being radically transformed. Curulla rightly notes that the religious dimension in theatre of the Revolution has been largely overlooked. While some scholars (Bérard, Brooks, Estève, and Marchand, for example) have broached the subject through specific plays, authors, or genres, Gender and Religious Life in French Revolutionary Drama is the first comprehensive study of the approximately two dozen extant plays "set in convents and monasteries" performed in Paris during the Revolution (10). For this reason and others, it is a welcome and important contribution to the field. Cogently argued, the book refutes some of the most persistent fallacies about the revolutionary stage, such as that all the pièces monacales were anticlerical propaganda. Highlighting the diversity of the monastic trope, Curulla offers contextualized readings of both well-known dramas (by Chénier, Monvel, and others) and lesser-known comedies, including Les Dragons et les Bénédictines and Les Visitandines. She is careful to make clear that stage depictions should be read not as realistic representations but rather as allegories that spoke to sociopolitical concerns of the time. Chapter 1 examines the popular (if largely mythic) figure of the unwilling nun forced to take religious vows by an unjust patriarch and corrupt clerics. Focusing on La Harpe's [End Page 259] sentimental drama Mélanie, ou la Religieuse about a novice who takes her life rather than take the veil, Curulla traces the play's fate from 1770 to 1803 to show how changing circumstances and the author's rewritings transformed its meaning. What began as a bold Enlightenment critique of the Church became a source of emotionally affecting performances in 1790, until the author disavowed it in 1803 after returning to Catholicism at the end of his life (though performances of the play continued). Although Mélanie is often cited by scholars of revolutionary theatre, Curulla's careful reading of the various scripts, contexts, and performances brings new insight, especially to the play's afterlife beyond 1794. In particular, she notes the juxtaposition of the chaste and hidden nuns in the play with the highly sexualized and visible actresses who played them, arguing that theatrical nuns were troubling gender prescriptions in the theatre as women's place in the new sociopolitical order was being publicly debated. Chapter 2 turns to several comedies in which nuns and monks leave the cloister to marry. Focusing especially on costume, Curulla reveals these stories of defrocked clerics to be vehicles for expressing new political and sexual identities in relation to changing concepts of citizenship. However, the popularity of these plays raised new fears: whereas initially the fear was that negative depictions of the religious onstage would diminish their status, by the Terror, the new fear was that sympathetic depictions of the clergy would improve their status with the public. Chapter 3 focuses on the conflation of...

‘Étranger,’ ou plutôt ‘fremd’: Philosophical-Poetic Nationalism in Derrida’s Geschlecht III and Beyond

This article takes up the specifically poetic dimension of what Jacques Derrida calls Martin Heidegger’s “philosophical nationalism” in the recently published Geschlecht III, arguing that this text doubles as a self-interrogation of Derrida’s own practice of reading poetry. Thus reading Geschlecht III alongside the nearly contemporaneous “Shibboleth: For Paul Celan,” I claim that Derrida’s critical deconstruction of Heidegger’s philosophical-poetic nationalism both allows us to read the traces of a more affirmatively deconstructive thinking of literary community in “Shibboleth” and draws attention to the limits and traps of such a project. Further, I demonstrate that Derrida’s and Heidegger’s respective approaches to the question of literary community cannot be separated from their respective approaches to the question of translation and their respective ways of mobilizing the motif of the “untranslatable.”

ACLSTM：A Novel Method for CQA Answer Quality Prediction Based on Question-answer Joint Learning

Given the limitations of the community question answering (CQA) answer quality prediction method in measuring the semantic information of the answer text, this paper proposes an answer quality prediction model based on the question-answer joint learning (ACLSTM). The attention mechanism is used to obtain the dependency relationship between the Question-and-Answer (Q&A) pairs. Convolutional Neural Network (CNN) and Long Short-term Memory Network (LSTM) are used to extract semantic features of Q&A pairs and calculate their matching degree. Besides, answer semantic representation is combined with other effective extended features as the input representation of the fully connected layer. Compared with other quality prediction models, the ACLSTM model can effectively improve the prediction effect of answer quality. In particular, the mediumquality answer prediction, and its prediction effect is improved after adding effective extended features. Experiments prove that after the ACLSTM model learning, the Q&A pairs can better measure the semantic match between each other, fully reflecting the model’s superior performance in the semantic information processing of the answer text.

Hierarchical Attentional Factorization Machines for Expert Recommendation in Community Question Answering

The most challenging task of Community Question Answering (CQA) is to provide high-quality answers to users’ questions. Currently, a variety of expert recommendation methods have been proposed and greatly improved the effective matching between questions and potential good answerers. However, the performance of existing methods can be adversely affected by many common factors such as data sparsity and noise problem, which cause less precise user modeling. Moreover, existing methods often model user-question interactions through simple ways, failing to capture the multiple scale interactions of question and answerers, which make it difficult to find answerers who are able to provide the best answers. In this paper, we propose an attention-based variant of Factorization Machines (FM) called Hierarchical Attentional Factorization Machines (HaFMRank) for answerer recommendation in CQA, which not only models the interactions between pairs of individual features but emphasizes the roles of crucial features and pairwise interactions. Specifically, we introduce the within-field attention layer to capture the inner structure of features belonging to the same field, while a feature-interaction attention layer is adopted to examine the importance of each pairwise interaction. A pre-training procedure is designed to generate latent FM feature embedding that encode question context and user history into the training process of HaFMRank. The performance of the proposed HaFMRank is evaluated by using real-world datasets of Stack Exchange and experimental results demonstrate that it outperforms several state-of-the-art methods in best answerer recommendation.

Forum Duplicate Question Detection by Domain Adaptive Semantic Matching

Community Question Answering (CQA) forums, such as Stack Overflow, Stack Exchange and Massive Open Online Course (MOOC) forums, spend a lot of manpower and time to manage duplicate questions on the forum. Mismatch of duplicate questions makes users keep asking “new” questions, and the continuous accumulation of duplicate questions may interfere with their information searching again, affecting user satisfaction. Neural Networks (NN) models for parsing semantics provide the possibility of end-to-end duplicate question detection. Whereas, due to lack of domain data and expertise, NN models for semantic parsing are rarely directly applied to CQA duplicate question detection. This paper proposes a Semantic Matching Model (SMM) integrated with the multi-task transfer learning framework for multi-domain forum duplicate question detection. By designing the word-to-sentence interaction mechanism based on the word-to-word interaction, SMM can automatically choose to ignore or pay attention to potential similar words according to the semantics at the sentence level. The experiments on the benchmark data set and MOOC forum data set state that SMM outperforms baselines, its interaction mechanism is effective and it has an advantage in cross-domain duplicate question detection.

Identifying Experts in Community Question Answering Website Based on Graph Convolutional Neural Network

High-quality answers, usually given by experts, play an important role in community question answering (CQA) websites. Therefore, experts in these websites are defined as those who provide high-quality answers. We propose two semi-supervised learning models based on the graph convolution neural network (GCN) to identify them. Both models comprehensively extract features from the social behavior network, user profiles, and question and answer text. Specifically, we construct a social behavior network according to the co-answering relationship among answerers, which means every two answerers are connected if they answer the same questions. The difference between these two models is the methods of extracting text features. One model named GCN-Doc uses Doc2vec to get text vectors before training. The other model named GCN-Lstm with a long short term memory (LSTM) network extracts text features while training. Experiments using real-world data from Zhihu.com, one of the largest Chinese CQA websites, show that both GCN-Doc and GCN-Lstm can identify experts effectively comparing with baselines of PageRank and other GCN based neural network models. Besides, GCN-Lstm performs better than GCN-Doc.

Dynamic user modeling for expert recommendation in community question answering

Community Question Answering (CQA) websites has become an important channel for people to acquire knowledge. In CQA, one key issue is to recommend users with high expertise and willingness to answer the given questions, i.e., expert recommendation. However, a lot of existing methods consider the expert recommendation problem in a static context, ignoring that the real-world CQA websites are dynamic, with users’ interest and expertise changing over time. Although some methods that utilize time information have been proposed, their performance improvement can be limited due to fact that they fail they fail to consider the dynamic change of both user interests and expertise. To solve these problems, we propose a deep learning based framework for expert recommendation to exploit user interest and expertise in a dynamic environment. For user interest, we leverage Long Short-Term Memory (LSTM) to model user’s short-term interest so as to capture the dynamic change of users’ interests. For user expertise, we design user expertise network, which leverages feedback on users’ historical behavior to estimate their expertise on new question. We propose two methods in user expertise network according to whether the dynamic property of expertise is considered. The experimental results on a large-scale dataset from a real-world CQA site demonstrate the superior performance of our method.

A Novel Bidirectional LSTM and Attention Mechanism based Neural Network for Answer Selection in Community Question Answering

A Novel Bidirectional LSTM and Attention Mechanism based Neural Network for Answer Selection in Community Question Answering

Hierarchy based firefly optimized K-means clustering for complex question answering

Complex Question Answering (CQA) is commonly used for answering community questions which requires human knowledge for answering them. It is essential to find complex question answering system for avoiding the complexities behind the question answering system. In the present work, we proposed Hierarchy based Firefly Optimized k-means Clustering (HFO-KC) method for complex question answering. Initially, the given input query is preprocessed. It eliminates the way of misclassification when comparing the strings. In order to enhance the answer selection process, the obtained keywords are mapped into the candidate solution. After mapping, the obtained keywords are segmented. Each segmentation forms a new query for answer selection and various number of answers selected for each query. Okapi-25 similarity computation is utilized for the process of document retrieval. Then the answers selected are classified with K means clustering which forms the hierarchy for each answer. Finally the firefly optimization algorithm is used for selecting the best quality of answer from the hierarchy.