基于内容的智能网络多媒体信息过滤检索

The paper discusses the construction of a content-based intelligent system that performs multimedia information filtering and retrieving on the Internet. The system disassembles the multimedia information into different media objects and describes them with vectors for content-based retrieval. In the user study module, the system uses the BP neural network to clarify the user interests for intelligent filtering and retrieving.     

Inferring user knowledge level from eye movement patterns

The acquisition of information and the search interaction process is influenced strongly by a person’s use of their knowledge of the domain and the task. In this paper we show that a user’s level of domain knowledge can be inferred from their interactive search behaviors without considering the content of queries or documents. A technique is presented to model a user’s information acquisition process during search using only measurements of eye movement patterns. In a user study (n = 40) of search in the domain of genomics, a representation of the participant’s domain knowledge was constructed using self-ratings of knowledge of genomics-related terms (n = 409). Cognitive effort features associated with reading eye movement patterns were calculated for each reading instance during the search tasks. The results show correlations between the cognitive effort due to reading and an individual’s level of domain knowledge. We construct exploratory regression models that suggest it is possible to build models that can make predictions of the user’s level of knowledge based on real-time measurements of eye movement patterns during a task session.     

Exploiting indoor location and mobile information for context-awareness service

Personal mobile devices such as cellular phones, smart phones and PMPs have advanced incredibly in the past decade. The mobile technologies make research on the life log and user-context awareness feasible. In other words, sensors in mobile devices can collect the variety of user’s information, and various works have been conducted using that information. Most of works used a user’s location information as the most useful clue to recognize the user context. However, the location information in the conventional works usually depends on a GPS receiver that has limited function, because it cannot localize a person in a building and thus lowers the performance of the user-context awareness. This paper develops a system to solve such problems and to infer a user’s hidden information more accurately using Bayesian network and indoor-location information. Also, this paper presents a new technique for localization in a building using a decision tree and signals for the Wireless LAN because the decision tree has many advantages which outweigh other localization techniques.     

Graph neural news recommendation with long-term and short-term interest modeling

With the information explosion of news articles, personalized news recommendation has become important for users to quickly find news that they are interested in. Existing methods on news recommendation mainly include collaborative filtering methods which rely on direct user-item interactions and content based methods which characterize the content of user reading history. Although these methods have achieved good performances, they still suffer from data sparse problem, since most of them fail to extensively exploit high-order structure information (similar users tend to read similar news articles) in news recommendation systems. In this paper, we propose to build a heterogeneous graph to explicitly model the interactions among users, news and latent topics. The incorporated topic information would help indicate a user’s interest and alleviate the sparsity of user-item interactions. Then we take advantage of graph neural networks to learn user and news representations that encode high-order structure information by propagating embeddings over the graph. The learned user embeddings with complete historic user clicks capture the users’ long-term interests. We also consider a user’s short-term interest using the recent reading history with an attention based LSTM model. Experimental results on real-world datasets show that our proposed model significantly outperforms state-of-the-art methods on news recommendation.     

Filtering search results using an optimal set of terms identified by an artificial neural network

Information filtering (IF) systems usually filter data items by correlating a set of terms representing the user’s interest (a user profile) with similar sets of terms representing the data items. Many techniques can be employed for constructing user profiles automatically, but they usually yield large sets of term. Various dimensionality-reduction techniques can be applied in order to reduce the number of terms in a user profile. We describe a new terms selection technique including a dimensionality-reduction mechanism which is based on the analysis of a trained artificial neural network (ANN) model. Its novel feature is the identification of an optimal set of terms that can classify correctly data items that are relevant to a user. The proposed technique was compared with the classical Rocchio algorithm. We found that when using all the distinct terms in the training set to train an ANN, the Rocchio algorithm outperforms the ANN based filtering system, but after applying the new dimensionality-reduction technique, leaving only an optimal set of terms, the improved ANN technique outperformed both the original ANN and the Rocchio algorithm.     

Multilingual sentence categorization and novelty mining

A challenge for sentence categorization and novelty mining is to detect not only when text is relevant to the user’s information need, but also when it contains something new which the user has not seen before. It involves two tasks that need to be solved. The first is identifying relevant sentences (categorization) and the second is identifying new information from those relevant sentences (novelty mining). Many previous studies of relevant sentence retrieval and novelty mining have been conducted on the English language, but few papers have addressed the problem of multilingual sentence categorization and novelty mining. This is an important issue in global business environments, where mining knowledge from text in a single language is not sufficient. In this paper, we perform the first task by categorizing Malay and Chinese sentences, then comparing their performances with that of English. Thereafter, we conduct novelty mining to identify the sentences with new information. Experimental results on TREC 2004 Novelty Track data show similar categorization performance on Malay and English sentences, which greatly outperform Chinese. In the second task, it is observed that we can achieve similar novelty mining results for all three languages, which indicates that our algorithm is suitable for novelty mining of multilingual sentences. In addition, after benchmarking our results with novelty mining without categorization, it is learnt that categorization is necessary for the successful performance of novelty mining.     

Integrating expert profile,reputation and link analysis for expert finding in question-answering websites

Question answering websites are becoming an ever more popular knowledge sharing platform. On such websites, people may ask any type of question and then wait for someone else to answer the question. However, in this manner, askers may not obtain correct answers from appropriate experts. Recently, various approaches have been proposed to automatically find experts in question answering websites. In this paper, we propose a novel hybrid approach to effectively find experts for the category of the target question in question answering websites. Our approach considers user subject relevance, user reputation and authority of a category in finding experts. A user’s subject relevance denotes the relevance of a user’s domain knowledge to the target question. A user’s reputation is derived from the user’s historical question-answering records, while user authority is derived from link analysis. Moreover, our proposed approach has been extended to develop a question dependent approach that considers the relevance of historical questions to the target question in deriving user domain knowledge, reputation and authority. We used a dataset obtained from Yahoo! Answer Taiwan to evaluate our approach. Our experiment results show that our proposed methods outperform other conventional methods.     

Probabilistic learning for selective dissemination of information

New methods and new systems are needed to filter or to selectively distribute the increasing volume of electronic information being produced nowadays. An effective information filtering system is one that provides the exact information that fulfills user's interests with the minimum effort by the user to describe it. Such a system will have to be adaptive to the user changing interest. In this paper we describe and evaluate a learning model for information filtering which is an adaptation of the generalized probabilistic model of Information Retrieval. The model is based on the concept of `uncertainty sampling', a technique that allows for relevance feedback both on relevant and nonrelevant documents. The proposed learning model is the core of a prototype information filtering system called ProFile.     

Collaborative filtering using orthogonal nonnegative matrix tri-factorization

Collaborative filtering aims at predicting a test user’s ratings for new items by integrating other like-minded users’ rating information. The key assumption is that users sharing the same ratings on past items tend to agree on new items. Traditional collaborative filtering methods can mainly be divided into two classes: memory-based and model-based. The memory-based approaches generally suffer from two fundamental problems: sparsity and scalability, and the model-based approaches usually cost too much on establishing a model and have many parameters to be tuned.     

Assessing user-specific difficulty of documents

On the web, a huge variety of text collections contain knowledge in different expertise domains, such as technology or medicine. The texts are written for different uses and thus for people having different levels of expertise on the domain. Texts intended for professionals may not be understandable at all by a lay person, and texts for lay people may not contain all the detailed information needed by a professional. Many information retrieval applications, such as search engines, would offer better user experience if they were able to select the text sources that best fit the expertise level of the user. In this article, we propose a novel approach for assessing the difficulty level of a document: our method assesses difficulty for each user separately. The method enables, for instance, offering information in a personalised manner based on the user’s knowledge of different domains. The method is based on the comparison of terms appearing in a document and terms known by the user. We present two ways to collect information about the terminology the user knows: by directly asking the users the difficulty of terms or, as a novel automatic approach, indirectly by analysing texts written by the users. We examine the applicability of the methodology with text documents in the medical domain. The results show that the method is able to distinguish between documents written for lay people and documents written for experts.     

User community detection via embedding of social network structure and temporal content

Identifying and extracting user communities is an important step towards understanding social network dynamics from a macro perspective. For this reason, the work in this paper explores various aspects related to the identification of user communities. To date, user community detection methods employ either explicit links between users (link analysis), or users’ topics of interest in posted content (content analysis), or in tandem. Little work has considered temporal evolution when identifying user communities in a way to group together those users who share not only similar topical interests but also similar temporal behavior towards their topics of interest. In this paper, we identify user communities through multimodal feature learning (embeddings). Our core contributions can be enumerated as (a) we propose a new method for learning neural embeddings for users based on their temporal content similarity; (b) we learn user embeddings based on their social network connections (links) through neural graph embeddings; (c) we systematically interpolate temporal content-based embeddings and social link-based embeddings to capture both social network connections and temporal content evolution for representing users, and (d) we systematically evaluate the quality of each embedding type in isolation and also when interpolated together and demonstrate their performance on a Twitter dataset under two different application scenarios, namely news recommendation and user prediction. We find that (1) content-based methods produce higher quality communities compared to link-based methods; (2) methods that consider temporal evolution of content, our proposed method in particular, show better performance compared to their non-temporal counter-parts; (3) communities that are produced when time is explicitly incorporated in user vector representations have higher quality than the ones produced when time is incorporated into a generative process, and finally (4) while link-based methods are weaker than content-based methods, their interpolation with content-based methods leads to improved quality of the identified communities.     

An approach to web-based Personal Health Records filtering using fuzzy prototypes and data quality criteria

Nowadays, new ways of managing and accessing to health-care information are continuously appearing. Web-based Personal Health Records (web PHRs) have the potential to make data about health-care available to clinicians, researchers and students in different medical contexts and applications. Therefore, the amount of web PHRs accessible through Internet has grown enormously and as a result health-care professionals are currently burdened with more and more data. It’s probable that these data, unfortunately, have not always the adequate levels of quality, making that their work cannot always be as successful as expected. As a way of alleviating this fact, the present work is focused on improving the document filtering results in the context of web PHRs management. To achieve this goal, a new kind of document filtering model is proposed. This model is based on fuzzy prototypes which are defined by means of conceptual prototypes. These prototypes are obtained by using a data quality analysis of documents. This analysis guarantees that filtered information will be relevant enough for the information user. The complete model provides an efficient strategy of document filtering that can be very useful when it is necessary to deal with a constant flow of new information.     

Aesthetics and credibility in web site design

Web sites often provide the first impression of an organization. For many organizations, web sites are crucial to ensure sales or to procure services within. When a person opens a web site, the first impression is probably made in a few seconds, and the user will either stay or move on to the next site on the basis of many factors. One of the factors that may influence users to stay or go is the page aesthetics. Another reason may involve a user’s judgment about the site’s credibility. This study explores the possible link between page aesthetics and a user’s judgment of the site’s credibility. Our findings indicate that when the same content is presented using different levels of aesthetic treatment, the content with a higher aesthetic treatment was judged as having higher credibility. We call this the amelioration effect of visual design and aesthetics on content credibility. Our study suggests that this effect is operational within the first few seconds in which a user views a web page. Given the same content, a higher aesthetic treatment will increase perceived credibility.     

Measuring the interestingness of articles in a limited user environment

Search engines, such as Google, assign scores to news articles based on their relevance to a query. However, not all relevant articles for the query may be interesting to a user. For example, if the article is old or yields little new information, the article would be uninteresting. Relevance scores do not take into account what makes an article interesting, which would vary from user to user. Although methods such as collaborative filtering have been shown to be effective in recommendation systems, in a limited user environment, there are not enough users that would make collaborative filtering effective.     

Intelligent agent systems for executive information scanning,filtering and interpretation: Perceptions and challenges

Using intelligent agent-based systems to support information processing for executives has not been significantly advanced in both theory and practice. Research into this field tends to focus more on technical aspects than on social perspective. When executives are faced with increasing information availability and uncertainty in the business environment, using intelligent agent-based systems to enhance executives’ information processing capability appears both an opportunity and a necessity. This study examines UK executives’ perceptions of intelligent agent-based systems for information scanning, filtering, interpretation and alerting. The study follows a deductive research design, i.e. hypothesis formulation and testing from the user’s perspective. Qualitative data was collected through focus groups and interviews with executives in the UK. The study produces rich evidence that challenges preconceptions of using agent-based information processing system by executives. The findings develop insight into executives’ behavior in information processing, which has implications for intelligent system developers and organizational information processing practice.     

Summarisation of the logical structure of XML documents

Summarisation is traditionally used to produce summaries of the textual contents of documents. In this paper, it is argued that summarisation methods can also be applied to the logical structure of XML documents. Structure summarisation selects the most important elements of the logical structure and ensures that the user’s attention is focused towards sections, subsections, etc. that are believed to be of particular interest. Structure summaries are shown to users as hierarchical tables of contents. This paper discusses methods for structure summarisation that use various features of XML elements in order to select document portions that a user’s attention should be focused to. An evaluation methodology for structure summarisation is also introduced and summarisation results using various summariser versions are presented and compared to one another. We show that data sets used in information retrieval evaluation can be used effectively in order to produce high quality (query independent) structure summaries. We also discuss the choice and effectiveness of particular summariser features with respect to several evaluation measures.     

A comparative survey of Personalised Information Retrieval and Adaptive Hypermedia techniques

A key driver for next generation web information retrieval systems is becoming the degree to which a user’s search and presentation experience is adapted to individual user properties and contexts of use. Over the past decades, two parallel threads of personalisation research have emerged, one originating in the document space in the area of Personalised Information Retrieval (PIR) and the other arising from the hypertext space in the field of Adaptive Hypermedia (AH).     

An evaluation of adaptive filtering in the context of realistic task-based information exploration

Exploratory search increasingly becomes an important research topic. Our interests focus on task-based information exploration, a specific type of exploratory search performed by a range of professional users, such as intelligence analysts. In this paper, we present an evaluation framework designed specifically for assessing and comparing performance of innovative information access tools created to support the work of intelligence analysts in the context of task-based information exploration. The motivation for the development of this framework came from our needs for testing systems in task-based information exploration, which cannot be satisfied by existing frameworks. The new framework is closely tied with the kind of tasks that intelligence analysts perform: complex, dynamic, and multiple facets and multiple stages. It views the user rather than the information system as the center of the evaluation, and examines how well users are served by the systems in their tasks. The evaluation framework examines the support of the systems at users’ major information access stages, such as information foraging and sense-making. The framework is accompanied by a reference test collection that has 18 tasks scenarios and corresponding passage-level ground truth annotations. To demonstrate the usage of the framework and the reference test collection, we present a specific evaluation study on CAFÉ, an adaptive filtering engine designed for supporting task-based information exploration. This study is a successful use case of the framework, and the study indeed revealed various aspects of the information systems and their roles in supporting task-based information exploration.