Distributed Representations of Expertise

Han, Fangqiu; Tan, Shulong; Sun, Huan; Srivatsa, Mudhakar; Yan, Xifeng

doi:10.1137/1.9781611974348.60

Cited by 11 publications

(14 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in the Table. I, we summarize the performance of these models on different types of matching tasks to explore the scope of application 3 . In the table, text VS text means to match text labels with text data, graph VS text means to match text labels with graph data, audio VS text is to match text labels with graph data; video VS text is to match text labels with video data.…”

Section: Categorization Of Expert Finding Techniquesmentioning

confidence: 99%

Expert finding in community question answering: a review

et al. 2019

View full text Add to dashboard Cite

The rapid development recently of Community Question Answering (CQA) satisfies users quest for professional and personal knowledge about anything. In CQA, one central issue is to find users with expertise and willingness to answer the given questions. Expert finding in CQA often exhibits very different challenges compared to traditional methods. Sparse data and new features violate fundamental assumptions of traditional recommendation systems. This paper focuses on reviewing and categorizing the current progress on expert finding in CQA. We classify all the existing solutions into four different categories: matrix factorization based models (MF-based models), gradient boosting tree based models (GBT-based models), deep learning based models (DL-based models) and ranking based models (R-based models). We find that MF-based models outperform other categories of models in the field of expert finding in CQA. Moreover, we use innovative diagrams to clarify several important concepts of ensemble learning, and find that ensemble models with several specific single models can further boosting the performance. Further, we compare the performance of different models on different types of matching tasks, including text vs. text, graph vs. text, audio vs. text and video vs. text. The results can help the model selection of expert finding in practice. Finally, we explore some potential future issues in expert finding research in CQA.

show abstract

Section: Categorization Of Expert Finding Techniquesmentioning

confidence: 99%

Expert finding in community question answering: a review

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The first stage selects the group with minimal distance to the ticket in embedding space. The group representations are learned using the model proposed by [30], using the tickets a group has resolved as positive samples and randomly selected negative samples. The second step follows the transition based model as proposed by [5], using group-group transition probabilities modeled in H T rans .…”

Section: Methodsmentioning

confidence: 99%

“…Agarwal et al [28] used POS tags together with syntactic features to evaluate ticket quality. After preprocessing, a ticket is converted to terms and can be represented in bag-ofwords models [30,31] and n-gram models [21]. Further, term weighting techniques, namely Term Frequency (TF) ( [8,32]) and Term Frequency Inverse Document Frequency (TF-IDF) ( [33][34][35][36]) are state-of-the-art in many text processing tasks, are applied in ticket processing.…”

Section: Ticket Information Extractionmentioning

confidence: 99%

“…The notion of Expert Network was first seen in the work of Shao et al [5], then adopted by [6,80,81]. The same concept is also referred to as Collaborative Network [30,31,82], Enterprise Network, or Resolution Network. We use the term "expert network" to differentiate this type of network involved in this study from other types of networks with identical names but for different purposes, e.g., "Collaborative Network" (i.e., social networks with collective efforts [83] and co-author networks [84]).…”

Section: Expert Network Analysismentioning

confidence: 99%

“…In the first stage, the initial group was determined by taking the nearest neighbor in a distributed expert group representation model, as per [30]. A ticket was projected to the same vector space as the expert groups for distance-based selection.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Mining software support tickets for assistive routing

Han¹

View full text Add to dashboard Cite

for sparing the time for discussions and reviewing my manuscripts. During my years studying in Nanyang Technological University of Singapore, I have the pleasure getting acquainted with many awesome people. Many of them are established researchers working in different fields, I wish them all the success and happiness in life.At SAP, I would like to thank my managers Steve Lee, Dr. Daniel Dahlmeier, and Dr. Chen Wang, for the careers guidance. A special regard to my first mentor Naveen Nandan, who introduced me to the company, and to Dr. Mohammad Akbari for his advice in research. For this special project on support tickets, I am always grateful to Andreas Heckmann, who envisioned an intelligent support application, and Christian Boos, who recommended me for the assignment. Since the beginning of the project, I worked closely with Project Manager Wilhelm Juette to form a productive team, his mentorship and friendship are always appreciated. Moreover, I wish to thank David Molinera, Kazuhiro Inaba and Ka Hian Goh for their hands-on support in the experiments.Most importantly, no word can describe my love and appreciation to my family, who are always there for me. Lastly, to my daughter. Learning is a lifelong journey, always be ready to embrace what is coming next.

show abstract

A Domain-Aware Crowdsourcing System with Copier Removal

Fang

Sun

et al. 2023

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

Distributed Representations of Expertise

Cited by 11 publications

References 15 publications

Expert finding in community question answering: a review

Expert finding in community question answering: a review

Mining software support tickets for assistive routing

A Domain-Aware Crowdsourcing System with Copier Removal

Contact Info

Product

Resources

About