语义万维网中基于Agent的智能检索框架

语义万维网是目前国际万维网联盟（world wide web consortium,W3C）为了解决因Web上的数据缺少语义信息而难以实现自动化处理的问题所开展的研究项目,其目的是为了对Web上发布的信息实现智能推理和自动化处理。Agent作为一种智能化主体,非常适合语义万维网环境下的各种应用。在语义万维网和智能Agent研究的基础上,综合信息检索、知识表示、Ontology建模等多方面技术,提出并实现了一个基于Ontology实现语义信息检索的多Agent系统框架,该系统包括信息收集、存储、查询和推理4个主要部分。     

A path-based approach for web page retrieval

Use of links to enhance page ranking has been widely studied. The underlying assumption is that links convey recommendations. Although this technique has been used successfully in global web search, it produces poor results for website search, because the majority of the links in a website are used to organize information and convey no recommendations. By distinguishing these two kinds of links, respectively for recommendation and information organization, this paper describes a path-based method for web page ranking. We define the Hierarchical Navigation Path (HNP) as a new resource for improving web search. HNP is composed of multi-step navigation information in visitors’ website browsing. It provides indications of the content of the destination page. We first classify the links inside a website. Then, the links for web page organization are exploited to construct the HNPs for each page. Finally, the PathRank algorithm is described for web page retrieval. The experiments show that our approach results in significant improvements over existing solutions.     

A unified framework for multimodal retrieval

In this paper, a unified framework for multimodal content retrieval is presented. The proposed framework supports retrieval of rich media objects as unified sets of different modalities (image, audio, 3D, video and text) by efficiently combining all monomodal heterogeneous similarities to a global one according to an automatic weighting scheme. Then, a multimodal space is constructed to capture the semantic correlations among multiple modalities. In contrast to existing techniques, the proposed method is also able to handle external multimodal queries, by embedding them to the already constructed multimodal space, following a space mapping procedure of a submanifold analysis. In our experiments with five real multimodal datasets, we show the superiority of the proposed approach against competitive methods.     

A novel optimization framework for salient object detection

Visual saliency aims to locate the noticeable regions or objects in an image. In this paper, a coarse-to-fine measure is developed to model visual saliency. In the proposed approach, we firstly use the contrast and center bias to generate an initial prior map. Then, we weight the initial prior map with boundary contrast to obtain the coarse saliency map. Finally, a novel optimization framework that combines the coarse saliency map, the boundary contrast and the smoothness prior is introduced with the intention of refining the map. Experiments on three public datasets demonstrate the effectiveness of the proposed method.     

A general shape context framework for object identification

A general shape context framework is proposed for object/image retrieval in occluded and cluttered environment with hundreds of models as the potential matches of an input. The approach is general since it does not require separation of input objects from complex background. It works by first extracting consistent and structurally unique local neighborhood information from inputs or models, and then voting on the optimal matches. Its performance degrades gracefully with respect to the amount of structural information that is being occluded or lost. The local neighborhood information applicable to the system can be shape, color, texture feature, etc. Currently, we employ shape information only. The mechanism of voting is based on a novel hyper cube based indexing structure, and driven by dynamic programming. The proposed concepts have been tested on database with thousands of images. Very encouraging results have been obtained.     

A framework for object migration in object-oriented databases

This article presents a framework developed for accomodating various object migrations in ‘statically-typed’ object databases. Requirements for supporting object migrations are stipulated, and a conceptual model for describing and facilitating different kinds of migrations is described. Associated issues of controlling such migrations are then addressed, along with an initial investigation on the interence of implied migration paths and the completeness of migration operators. Some guidelines are then given to help users conduct migrations more effectively. An implementation prototype on top of an object-oriented database system was built, which embodies full support of all migration types specified in the migration model.     

A genetic programming framework for content-based image retrieval

The effectiveness of content-based image retrieval (CBIR) systems can be improved by combining image features or by weighting image similarities, as computed from multiple feature vectors. However, feature combination do not make sense always and the combined similarity function can be more complex than weight-based functions to better satisfy the users’ expectations. We address this problem by presenting a Genetic Programming framework to the design of combined similarity functions. Our method allows nonlinear combination of image similarities and is validated through several experiments, where the images are retrieved based on the shape of their objects. Experimental results demonstrate that the GP framework is suitable for the design of effective combinations functions.     

CACHE RP : A novel dynamic cache size tuning model working with relative object popularity for fast web information retrieval

The CACHE_RPframework leverages the relative object popularity as the sole parameter for dynamic cache size tuning. In the process it consistently maintains the prescribed cache hit ratio on the fly by deriving the popularity ratio from the current statistics. As a result the accuracy of the statistical CACHE_RPoperation is independent of changes in the Internet traffic pattern that may switch suddenly. The contribution by the novel CACHE_RPframework is that by adaptively maintaining the given hit ratio it effectively reduces the end-to-end information retrieval roundtrip time (RTT) and frees more bandwidth for sharing. This bandwidth would otherwise be consumed in transferring large amounts of data from remote data sources to the proxy server before it is given to the client or requestor.     

Cyberbroker: A lightweight web object collector

Centralized heavyweight collectors have been used to collect text-based web objects for building search engines. However, these collectors do not meet the need of collecting big web objects such as images and videos. We propose a lightweight mobile on-call object collector for the world wide web. Using this approach, we can deploy a collector easily on the site being indexed. We present a prototype system consisting of mobile Java collectors and search engines for collecting HTML/image objects. We provide analysis to find a lower bound in terms of total data transfer time required. This idealized scenario is complemented with another model based on implementation of the HTTP over TCP. The lower bound and the upper bound of the gains accrued are then validated using experiments. This revised version was published online in August 2006 with corrections to the Cover Date.     

A framework for corroborating answers from multiple web sources

Search engines are increasingly efficient at identifying the best sources for any given keyword query, and are often able to identify the answer within the sources. Unfortunately, many web sources are not trustworthy, because of erroneous, misleading, biased, or outdated information. In many cases, users are not satisfied with the results from any single source. In this paper, we propose a framework to aggregate query results from different sources in order to save users the hassle of individually checking query-related web sites to corroborate answers. To return the best answers to the users, we assign a score to each individual answer by taking into account the number, relevance and originality of the sources reporting the answer, as well as the prominence of the answer within the sources, and aggregate the scores of similar answers. We conducted extensive qualitative and quantitative experiments of our corroboration techniques on queries extracted from the TREC Question Answering track and from a log of real web search engine queries. Our results show that taking into account the quality of web pages and answers extracted from the pages in a corroborative way results in the identification of a correct answer for a majority of queries.     

A cost driven disk scheduling algorithm for multimedia object retrieval

This paper describes a novel cost-driven disk scheduling algorithm for environments consisting of multipriority requests. An example application is a video-on-demand (VOD) system that provides high and low quality services, termed priority 2 and 1, respectively. Customers ordering a high quality (priority 2) service pay a higher fee and are assigned a higher priority by the underlying system. Our proposed algorithm minimizes costs by maintaining one-queue and managing requests intelligently in order to meet the deadline of as many priority 1 requests as possible while maximizing the number of priority 2 requests that meet their deadline. Our algorithm is general enough to accommodate an arbitrary number of priority levels. Prior schemes, collectively termed "multiqueue" schemes maintain a separate queue for each priority level in order to optimize the performance of the high priority requests only. When compared with our proposed scheme, in certain cases, our technique provides more than one order of magnitude improvement in total cost.     

A framework for visual-context-aware object detection in still images

Visual context provides cues about an object’s presence, position and size within the observed scene, which should be used to increase the performance of object detection techniques. However, in computer vision, object detectors typically ignore this information. We therefore present a framework for visual-context-aware object detection. Methods for extracting visual contextual information from still images are proposed, which are then used to calculate a prior for object detection. The concept is based on a sparse coding of contextual features, which are based on geometry and texture. In addition, bottom-up saliency and object co-occurrences are exploited, to define auxiliary visual context. To integrate the individual contextual cues with a local appearance-based object detector, a fully probabilistic framework is established. In contrast to other methods, our integration is based on modeling the underlying conditional probabilities between the different cues, which is done via kernel density estimation. This integration is a crucial part of the framework which is demonstrated within the detailed evaluation. Our method is evaluated using a novel demanding image data set and compared to a state-of-the-art method for context-aware object detection. An in-depth analysis is given discussing the contributions of the individual contextual cues and the limitations of visual context for object detection.