研究
情報検索¶
Text/Image Retrieval¶
The target of this research is to develop enhanced techniques to improve the user-experience when users are browsing, searching and retrieving documents/images.
Privacy-preserving Information Retrieval¶
Modern IR systems can achieve enhanced performance by analyzing huge amounts of log data gathered from users. Unfortunately, the data to derive such insights is personal and sensitive, which might give rise to catastrophic consequences, even if the system collecting such data has resolved to ‘do no evil’. By now, accurately and efficiently providing satisfactory results to users while preserving privacy is far from being resolved. This research topic aims to initiate research into privacy-preserving IR and develop a scalable privacy-preserving IR system.
Neural Information Retrieval¶
The application of deep learning has attracted great attention as a breakthrough and yielded the state-of-the-art performance in many tasks. This research is to make an in-depth exploration of the utilization of deep neural networks models in the field of Information Retrieval.
Diversified Information Retrieval¶
To satisfy users with different information needs, this technique aims to provide a diversified search result, which features a trade-off between relevance and diversity. This technique is widely used in many fields, such as web search and recommendation systems.
Interactive Information Retrieval¶
The technique of Interactive Information Retrieval allows user interaction and provides adaptive search results. Essentially it can be viewed as context-driven information retrieval, the context information includes previously submitted queries, interactive behaviors, etc. Fine-grained topics include user modeling, intent identification, behavior understanding, adaptive item ranking, etc.
Design of Evaluation Metric¶
This techqiue aims to quantify and compare the effectiveness of different information retrieval methods. The main tasks include designing evaluation metrics, collecting data, building test collections, etc.
ユーザー理解¶
User Modeling¶
Nowadays, enormous volume of search requests are submitted everyday. For example, Google processes over 3.5 billion searches per day (according to Internet Live Stats). Query logs capture and store interactions between search engines and their users and comprise a source of rich information regarding the ways in which users express their information needs, seek and select desired information units.
This technique aims to extract the knowledge embedded in query logs in order to understand users, facilitate research of relevant fields, such as Information Retrieval and Image Understanding.
User Intent Identification¶
Given an input query, user intent identification is the challenging problem of predicting the possible information needs.
画像理解¶
It refers to the subdiscipline of Artificial Intelligence (AI) that tries to make the computers see.
Image Caption Generation¶
Given an image, caption generation is the challenging problem of generating a human-readable textual description.
Metric Learning¶
Metric learning is the task of learning a distance function over images.
知識グラフ¶
Virtual Knowledge Graph¶
Different from the traditional knowledge graph, such as Freebase, DBpedia, Yago and Wikidata, virtual knowledge graph directly treats a large-scale corpus as a knowledge base. In short, the technique of virtual knowledge graph aims to provide open-ended commonsense knowledge given a raw corpus.
Behavioral Knowledge Graph¶
This research topic explores : (1) how to effectively and efficiently build a large-scale behavioral knowledge graph which encodes rich action-related information; (2) how to deploy behavioral knowledge graph into real applications.