Titlebook: Information Retrieval Technology; 6th Asia Information Pu-Jen Cheng,Min-Yen Kan,Preslav Nakov Conference proceedings 2010 Springer Berlin H

Vulnerary

Revisiting Rocchio’s Relevance Feedback Algorithm for Probabilistic Modelsfeedback model in the language modeling framework. Moreover, the QRocDFR method considerably improves the robustness of RocDFR’s retrieval performance with respect to the size of feedback document set.

爆炸

The Role of Lexical Ontology in Expanding the Semantic Textual Content of On-Line News Images named entity recognition (NER) technique to annotate important entities such as name if a person, location and organization among image searchers. Results show that lexical ontology has the capacity to semantically enhance the meanings of conventional bag of words index.

motivate

昏睡中

Score Estimation, Incomplete Judgments, and Significance Testing in IR Evaluationones, admitting a degree of uncertainty into the per-topic effectiveness scores being compared. Here we explore the design options that must be considered when planning such an experimental evaluation, with emphasis on how effectiveness scores are inferred from partial information.

GOUGE

Effective Time Ratio: A Measure for Web Search Engines with Document Snippetsto .. For search engine with snippet, some theoretical analysis proves that its value can reflect both retrieval system performance and snippet quality. We further deploy a real user study, showing that . can reflect users’ satisfaction better than the existing metrics based on document relevance and/or snippet relevance.

FATAL

显赫的人

POWER

Relevance Ranking Using Kernelsh to relevance ranking using the kernel technique in statistical learning. In the approach, the general ranking model is defined as a kernel function of query and document representations. A number of kernel functions are proposed as specific ranking models in the paper, including BM25 Kernel, LMIR

Immunization

obviate

Title-Based Product Search – Exemplified in a Chinese E-commerce Portalsystem seems suitable for searching products. Unfortunately, we observe from real world query logs and find that queries for product search are usually very short. What is worse, a document described a product may have lots of words of related products. It is hard for an IR system to distinguish rep