Information Retrieval Technology

Lecture notes in computer sciencePublished 1 January 2016

Zhao, Xin, Yi Chang, Min Zhang, Zhicheng Dou, Yiqun Liu, Ji-Rong Wen

DOI10.1007/978-3-319-48051-0

Source URL

Citations78

SJR quartileQ2

SJR score0.35

SNIP0.55

TL;DR

This work investigates the relationship between relevance likelihood and retrieval rank, seeking to identify plausible methods for estimating document relevance and hence computing an inferred gain.

Abstract

This book constitutes the refereed proceedings of the 12th Information Retrieval Societies Conference, AIRS 2016, held in Beijing, China, in November/December 2016. The 21 full papers presented together with 11 short papers were carefully reviewed and selected from 57 submissions. The final programme of AIRS 2015 is divided in the following tracks: IR models and theories; machine learning and data mining for IR; IR applications and user modeling; personalization and recommendation; and IR evaluation.

Keywords

Computer Science

Lecture notes in computer scienceQuery Subtopic Mining Exploiting Word Embedding for Search Result Diversification

10 Citations2016Md Zia Ullah, Md Shajalal +2 more

This paper proposes mining query subtopic by exploiting the word embedding and short-text similarity measure and introduces a new way of ranking based on a new novelty estimation that faithfully represents the possible search intents of the query.

Lecture notes in computer scienceWhen MetaMap Meets Social Media in Healthcare: Are the Word Labels Correct?

5 Citations2016Hongkui Tu, Zongyang Ma +2 more

This study manually annotates the correctness of medical words extracted by MetaMap from 100 posts in HealthBoards and gets a precision of 43.75 %.

Lecture notes in computer sciencePredicting Information Diffusion in Social Networks with Users’ Social Roles and Topic Interests

4 Citations2016Xiaoxuan Ren, Yan Zhang

An approach, Role and Topic aware Independent Cascade (RTIC), to uncover information diffusion in social networks, which extracts the opinion leaders and structural hole spanners and analyze the users’ interests on specific topics is proposed.

Lecture notes in computer scienceLearning to Improve Affinity Ranking for Diversity Search

4 Citations2016Yue Wu, Jingfei Li +2 more

The proposed learning model not only considers the topic coverage maximization and redundancy reduction by formalizing a series of features, but also optimizes the diversity metric by extending a well-known learning-to-rank algorithm LambdaMART.

Lecture notes in computer scienceEvaluation with Confusable Ground Truth

2 Citations2016Jiyi Li, Masatoshi Yoshikawa

A term of confusability is defined which represents how confusable the reviewers are on the instances and can be used as a supplementary to existing rating aggregation methods and evaluation methods.

Lecture notes in computer scienceLearning to Rank with Likelihood Loss Functions

1 Citations2016Yuan Lin, Liang Yang +3 more

A novel machine learning framework for ranking, which depends on two new samples, and a novel preference-weighted loss function for multiple relevance judgment data sets, which are effective in improving ranking performance.

Lecture notes in computer scienceAssessing the Authors of Online Books in Digital Libraries Using Users Affinity

1 Citations2016Baptiste de La Robertie

The problem of assessing the authors of users generated published books in digital libraries is addressed and it is proposed to model the platform using an heterogeneous graph representation and to exploit both the users’ interests and the natural inter-users affinities to infer the author of unlabelled books.

Lecture notes in computer scienceAn In-Depth Study of Implicit Search Result Diversification

1 Citations2016Hai-Tao Yu, Adam Jatowt +5 more

Experimental results show that ILP4ID can achieve substantially improved performance over the state-of-the-art methods in terms of standard diversity metrics.

Lecture notes in computer scienceReformulate or Quit: Predicting User Abandonment in Ideal Sessions

1 Citations2016Mustafa Zengin, Ben Carterette

It is shown that under ideal conditions for identifying topical sessions, the best features are those related to user actions and document relevance, while features related to query/document similarity actually hurt prediction abandonment.