Joint Geo-Spatial Preference and Pairwise Ranking for Point-of-Interest Recommendation
- Fajie Yuan, J. Jose, G. Guo, Long Chen, Haitao Yu, Rami Suleiman Alkhawaldeh
- Computer ScienceIEEE International Conference on Tools with…
- 1 November 2016
A co-pairwise ranking model based on the assumption that users prefer to assign higher ranks to the POIs near previously rated ones is proposed, which can learn preference ordering from non-observed rating pairs, and thus can alleviate the sparsity problem of matrix factorization.
LambdaFM: Learning Optimal Ranking with Factorization Machines Using Lambda Surrogates
- Fajie Yuan, G. Guo, J. Jose, Long Chen, Haitao Yu, Weinan Zhang
- Computer ScienceInternational Conference on Information and…
- 24 October 2016
This paper introduces Lambda Factorization Machines (LambdaFM), which is particularly intended for optimizing ranking performance for IFCAR, and creates three effective lambda surrogates by conducting a theoretical analysis for lambda from the top-N optimization perspective.
Overview of NTCIR-12 Temporal Information Access (Temporalia-2) Task
- Hideo Joho, A. Jatowt, Roi Blanco, Haitao Yu, Shuhei Yamamoto
- Computer ScienceNTCIR Conference on Evaluation of Information…
- 2016
This paper overviews NTCIR-12 Temporal Information Access (Temporalia-2) task, which aims to foster research in temporal aspects of information retrieval and search and describes both the subtasks, datasets, evaluation methods and the results of meta analyses.
Overview of NTCIR-13 Actionable Knowledge Graph (AKG) Task
- Roi Blanco, Hideo Joho, A. Jatowt, Haitao Yu, Shuhei Yamamoto
- Computer ScienceNTCIR Conference on Evaluation of Information…
- 2017
This paper overviews NTCIR-13 Actionable Knowledge Graph (AKG) task. The task focuses on finding possible actions related to input entities and the relevant properties of such actions. AKG is…
BoostFM: Boosted Factorization Machines for Top-N Feature-based Recommendation
- Fajie Yuan, G. Guo, J. Jose, Long Chen, Haitao Yu, Weinan Zhang
- Computer ScienceInternational Conference on Intelligent User…
- 7 March 2017
BoostFM is an adaptive boosting framework that linearly combines multiple homogeneous component recommenders which are repeatedly constructed on the basis of the individual FM model by a re-weighting scheme, which outperforms a number of state-of-the-art approaches for top-N recommendation.
Mining Large-scale Comparable Corpora from Chinese-English News Collections
- Degen Huang, Lian Zhao, Lishuang Li, Haitao Yu
- Computer Science, EducationInternational Conference on Computational…
- 23 August 2010
A CLIR-based approach to construct large-scale Chinese-English comparable corpora, which is valuable for translation knowledge mining, and experimental results indicate that this approach is effective on the construction of Chinese- English comparable Corpora.
Search Result Diversification via Filling Up Multiple Knapsacks
The experimental results show that the proposed 0-1 MSKP model outperforms several state-of-the-art methods significantly, not only in terms of standard diversity metrics (α-nDCG, nERRIA and subtopic recall), but also in termsof efficiency.
TUTA1 at the NTCIR-11 Temporalia Task
In the NTCIR-11 Temporalia task including Temporal Query Intent Classication (TQIC) and Temporal Information Retrieval, semi-supervised and supervised linear classiers are learned to predict the temporal classes for each search query.
Optimize What You Evaluate With: Search Result Diversification Based on Metric Optimization
- Haitao Yu
- Computer ScienceAAAI Conference on Artificial Intelligence
- 28 June 2022
A novel framework through direct metric optimization for SRD based on the score-and-sort strategy that jointly scores candidate documents by taking into account both cross-document interaction and permutation equivariance, which makes it possible to generate a diversified ranking via a simple sorting.
Topic detection and tracking on heterogeneous information
- Long Chen, Huaizhi Zhang, J. Jose, Haitao Yu, Yashar Moshfeghi, P. Triantafillou
- Computer ScienceJournal of Intelligence and Information Systems
- 1 August 2018
This work implemented a heterogeneous topic model that enables topic–topic correspondence between the sources by iteratively updating its topic–word distribution and captures temporal dynamics of topics from heterogeneous sources by exploiting both their individual properties and their inter-relationships.
...
...