Abstract: Semantic annotation is at the core of Semantic Web technology: it bridges the gap between legacy non-semantic web resource descriptions and their elicited, formally specified conceptualization, converting syntactic structures into knowledge structures, i.e., ontologies. Most existing approaches and tools are designed to deal with manual or semi-/automatic semantic annotation that exploits available ontologies through the pattern-based discovery of concepts. This work aims to generate the automatic semantic annotation of web resources, without any prefixed ontological support. The novelty of our approach is that, starting from web resources, content with a high-level of abstraction is obtained: concepts, connections between concepts, and instance-population are identified and arranged into an ex-novo ontology. The framework is designed to process resources from different sources (textual information, images, etc.) and generate an ontology-based annotation. A data-driven analysis reveals the data and their intrinsic relationships (in the form of triples) extracted from the resource content. On the basis of the discovered semantics, corresponding concepts and properties are modeled, allowing an ad hoc ontology to be built through an OWL-based coding annotation. The benefit of this approach is the generation of knowledge structured in a quite automatic way (i.e., the human support is restricted to the configuration of some parameters). The approach exploits a fuzzy extension of the mathematical modeling of Formal Concept Analysis and Relational Concept Analysis to generate the ontological structure of data resources.

Abstract: The web encourages the constant creation and distribution of large amounts of information; it is also a valuable resource for understanding human behavior and communication. To take full advantage of the web as a research resource that extends beyond the consideration of snapshots of the present, however, it is necessary to begin to take web archiving much more seriously as an important element of any research program involving web resources. The ephemeral character of the web requires that researchers take proactive steps in the present to enable future analysis. Efforts to archive the web or portions thereof have been developed around the world, but these efforts have not yet provided reliable and scalable solutions. This article summarizes the current state of web archiving in relation to researchers and research needs. Interviews with researchers, archivists, and technologists identify the differences in purpose, scope, and scale of current web archiving practice, and the professional tensions that arise given these differences. Findings outline the challenges that still face researchers who wish to engage seriously with web content as an object of research, and archivists who must strike a balance reflecting a range of user needs.

Abstract: This paper seeks to investigate credibility, misinformation and disinformation as concepts highly correlated to the quality of information sources so as to encourage users to bear them in mind when searching for information via the web. Issues as to how users can make distinction among web information sources when confronting questionable ones are discussed. Exploring within an extensive, but not comprehensive, body of works related to the main issues, the paper attempts to integrate t hem into a conceptual framework and even to find criteria by which web resources could be evaluated. Using some information skills like checklists, critical thinking and information literacy, users can considerably lessen challenges posed by searching for credible information from the web. Aside from novelty of the two concepts misinformation and disinformation in the information behavior literature, the paper provides a framework for the concepts explored and a list of proposed solutions by which credibility of web information could be assessed. © 2014 Infopreneurship Journal

Abstract: Most archived HTML pages embed other web resources, such as images and stylesheets. Playback of the archived web pages typically provides only the capture date (or Memento-Datetime) of the root resource and not the Memento-Datetime of the embedded resources. In the course of our research, we have discovered that the Memento-Datetime of embedded resources can be up to several years in the future or past, relative to the Memento-Datetime of the embedding root resource. We introduce a framework for assessing temporal coherence between a root resource and its embedded resource depending on Memento-Datetime, Last-Modified datetime, and entity body.

Abstract: As the today's Web provides open communication environment for a variety of web resources, the Web of Things (WoT) offers new opportunity and challenges about the interoperation among the smart things. The well-known Web technologies can leverage the Web-enabled things to publish and exchange their resource information over the Web, then the Web-enabled thing should cope with the security threat regarding the information exposures over the Web, particularly, access permissions to the thing's resource information. Thus, in this paper we analyse access permission control mechanism considering both the WoT characteristics and the REST-compliant resource-oriented Web architecture. In contrast to existing access control logics, the proposed mechanism utilizes not only the requester information such as the typical identity and the internet addresses, but also the context of the thing itself. Based on this mechanism, we present web-resource structure for access permission control, and describe an exemplary procedure in detail. This research contributes to the flexible and decentralized access permission control for WoT.

Abstract: Many Web portals allow users to associate additional information with existing multimedia resources such as images, audio, and video. However, these portals are usually closed systems and user-generated annotations are almost always kept locked up and remain inaccessible to the Web of Data. We believe that an important step to take is the integration of multimedia annotations and the Linked Data principles. We present the current state of the Open Annotation Model, explain our design rationale, and describe how the model can represent user annotations on multimedia Web resources. Applying this model in Web portals and devices, which support user annotations, should allow clients to easily publish and consume, thus exchange annotations on multimedia Web resources via common Web standards.

Abstract: The 2014 Web Server Issue of Nucleic Acids Research is the 12th in a series of annual special issues dedicated to web-based software resources for analysis and visualization of molecular biology data. It is freely available online under NAR's open access policy. The present issue reports on 80 web servers. Topics. This year's special emphasis is on tools for synthetic biology design, analysis of high throughput sequencing data, network and pathway analysis, and biological text mining. A total of 19 papers deal with these topics. The other major topic categories include analysis involving DNA and RNA (12 papers); gene prioritization (5 papers); drug-target interactions (5 papers); and proteins, including structure, ligand binding, and functional site prediction (24 papers). The 2014 Web Server issue continues the presentation of two special categories, one for stand-alone programs that analyze high-throughput data, such as high throughput sequencing data, and one for large collections of web services for automated analyses that can be utilized programmatically rather than through manual interaction with a web browser. Six papers fall in these categories. Acknowledgements. The Web Server issue would not be possible without the work of the many scientists and programmers who have provided us with outstanding, freely available web resources, and the conscientious efforts of literally hundreds of reviewers. My work was made possible, first, by the editorial assistance of Fay Oppenheim. Thank you. Thanks also to Allyson Byrd and Joe Perez-Rogers, PhD students in the Boston University Bioinformatics Program, and Artem Mamonov, Research Associate in the BU department of Biomedical Engineering, for their outstanding assistance in evaluating the proposal websites (Figure 1). Additional thanks to Martine Bernardes-Silva, Editorial Manager, NAR, and Jennifer Boyd and the staff at Oxford University Press. Instructions for Submissions. To streamline the review process, authors are required to send a one-page summary of their web server to the editor, Dr. Gary Benson (ude.ub@vrsbwran), for pre-approval prior to manuscript submission. For the 2014 issue, 269 summaries were submitted and 106, or 39%, were approved for manuscript submission. Of those approved, 80, or 75%, were ultimately accepted for publication. Review of a summary includes evaluation of the proposal and extensive testing of web server functionality. The key criteria for pre-approval are high scientific quality, wide interest, the ability to do computations on user-submitted data, and a well designed, well implemented, and fully functional website. Note that there is a minimum two-year interval before publication in the Web Server issue for web servers, or essentially similar web servers, that have been the subject of a previous publication, including publication in journals other than NAR. With respect to the website, the following are guidelines for approval. It should have an easy-to-find submission page with a simple mechanism for loading test data and setting test parameters. The preferred method is one-click loading. Additional mechanisms that assist the user in submitting data should be implemented where appropriate, for example, automatic download of a pdb structure file once the user has entered the appropriate identifier. Output should be dynamic and rich in detail. Wherever possible, supporting evidence used in calculations and/or links to external databases containing additional information should be provided. Numerical, textual, and visual output should be mixed and any visualization tools that add information or increase the user's understanding should be utilized. Note that output consisting merely of a few numerical values, a static spreadsheet, or a series of files to be opened in other programs will not be approved. Web servers that do not finish their calculations immediately must implement a mechanism for returning results to the user. Notification by email may be provided as an option, but an alternative that returns a web link at the time of data submission, which the user can then bookmark and access at a later time, is required. This link should ideally report the status of the job (queued, running, finished). Websites that require a guest login will not be approved. Note that uploaded data and the results of analysis for each user must be private and not viewable by other users. The website should be supported by an extensive help section or tutorial that guides the user through the submission process, contains details about input file formats and parameters, and importantly, explains the meaning of the output. Whenever possible, the help pages should link to dynamic output examples similar to those provided by the website. Text and figure help pages, rather than video tutorials, are preferred because they simplify quick look-up. Any proposal for a web server that is predictive must include details on validation of predictions from new data not used in training. N-fold cross validation methods will not be considered sufficient. Details should include size and composition of the validation data set (number of positive and negative cases), and several measures of predictive performance, including sensitivity, specificity, and precision. Proposals are frequently rejected for lack of adequate prediction validation information. Websites not clearly designed to accept and analyze user-submitted data will be rejected. This applies to those established primarily for lookup or exploration in a data set, or serve the function of “data aggregators.” Authors of websites that provide novel data should consider the NAR Database Issue as a possible venue (see the instructions at http://www.oxfordjournals.org/our_journals/nar/for_authors/msprep_database.html). Proposals that describe a new analysis method are generally not appropriate for the Web Server issue because limited space makes adequate method description and validation problematic. Authors of such methods might instead consider sending their manuscript to NAR as a regular computational biology paper (see the instructions for authors at http://www.oxfordjournals.org/our_journals/nar/for_authors/criteria_scope.html#Computational%20Biology). Figure 1. Allyson Byrd and Joe Perez-Rogers, PhD students in the Boston University Bioinformatics Graduate Program, and Artem Mamonov, Research Associate in the BU Department of Biomedical Engineering, helped with testing the Web Server proposal websites. Special Emphasis for 2015. For the 2015 issue, the topics of special emphasis will be tools for synthetic biology design, analysis of high throughput sequencing data, and network and pathway analysis. Deadlines for 2015. Authors wishing to submit manuscripts for the 2015 Web Server issue must submit their one page proposal along with the URL address of the fully functional website to ude.ub@vrsbwran by 31 December 2014. Detailed instructions and requirements are presented at http://www.oxfordjournals.org/nar/for_authors/submission_webserver.html. This information should be consulted before sending in the summary. The deadline for submission of articles is 31 January 2015. Requirement for References Links. Manuscripts submitted for the 2015 issue must format their References section to include active links to electronic versions of the cited papers, including links to PubMed, PubMed Central, and a DOI link. Instructions for incorporating these links into the manuscript are presented at http://www.oxfordjournals.org/nar/for_authors/submission_webserver.html.

Abstract: One of the challenges facing the current web is the efficient use of all the available information. The Web 2.0 phenomenon has favored the creation of contents by average users, and thus the amount of information that can be found for diverse topics has grown exponentially in the last years. Initiatives such as linked data are helping to build the Semantic Web, in which a set of standards are proposed for the exchange of data among heterogeneous systems. However, these standards are sometimes not used, and there are still plenty of websites that require naive techniques to discover their contents and services. This paper proposes an integrated framework for content and service discovery and extraction. The framework is divided into several layers where the discovery of contents and services is made in a representational stateless transfer system such as the web. It employs several web mining techniques as well as feature-oriented modeling for the discovery of cross-cutting features in web resources. The framework is used in a scenario of electronic newspapers. An intelligent agent crawls the web for related news, and uses services and visits links automatically according to its goal. This scenario illustrates how the discovery is made at different levels and how the use of semantics helps implement an agent that performs high-level tasks.

Abstract: A key challenge of the Semantic Web lies in the creation of semantic links between Web resources. The creation of links serves as a mean to semantically enrich Web resources, connecting disparate information sources and facilitating data reuse and sharing. As the amount of data on the Web is ever increasing, automated methods to unveil links between Web resources are required. In this paper, we introduce a tool, called SCS Connector, that assists users to uncover links between entity pairs within and across datasets. SCS Connector provides a Web-based user interface and a RESTful API that enable users to interactively visualise and analyse paths between an entity pair \((e_i,e_j)\) through known links that can reveal meaningful relationships between \((e_i,e_j)\) according to a semantic connectivity score (\(SCS\)).

Abstract: In recent years, the internet has emerged as the most important and powerful medium for the communication of information. There is a tremendous growth in the number and variety of information resources available on the internet which becomes an important source for scholarly scientific literature and also more number of information resources as well as the results of scientific and medical research is now being available on web. The paper describes the use of web resources (e-journals and e-databases subscribed by UGC-Info-net consortium) by the students of medical sciences at Aligarh Muslim University, India. A well structured questionnaire was administered to 120 students to collect the primary data from respondents. A total number of 92 filled in questionnaires were received showing overall response rate of 76.66 %. The paper also indicates that it is probably counter-productive to evaluate students as one group. Different segments of students have very different and varied use patterns of web resources depending on study topic, study year, psychological dispositions, and other demographic factors. DESIDOC Journal of Library & Information Technology , 2014, 34(2), pp. 125-130. http://dx.doi.org/10.14429/djlit.34.4006

Abstract: A huge amount of heterogeneous learning materials are generated on the web everyday with the rapid increase in the development of online learning technology. Besides, the learning resources are growing infinitely making it difficult for users to choose appropriate resources for their learning. Recommender systems, a subset of information filtering shows a great potential to help users in a personal learning environment to identify relevant and interesting items from a large number of items by suggesting actions to a user based on the preferences and ratings of other learners. The recommendation could be an online activity, running an online simulation or just a simple web resource. The technology finds its applications in a wide range of fields such as movies, music, news social tags, research articles, experts, social tags ,products, restaurants, jokes, financial services etc., This paper reviews the main paradigms of recommender systems and also the various methodologies that have been implemented to design recommender systems for personal learning environments.

Abstract: Emerging Semantic Web technology offered the Resource Description Framework (RDF) as a standard for semantic annotation of Web resources. It is expected that Web content with RDF-based metadata layer and ontological basis for it will be enough to enable interoperable and automated processing of Web data by various applications. However emerging industrial applications consider e.g. machines, processes, personnel, services for condition monitoring, remote diagnostics and maintenance, etc. to be specific classes of Web resources and thus a subject for semantic annotation. Such resources are naturally dynamic, not only from the point of view of changing values for some attributes (state of resource) but also from the point of view of changing “status-labels” (condition of the resource). In this paper we present Resource State/Condition Description Framework (RSCDF), as an extension to RDF, which introduces upper-ontology for describing such characteristics of resources as states and correspondent conditions, dynamics of state changes, target conditions and historical data about previous states. These descriptions are supposed to be used by external Web-services (e.g. condition monitoring, remote diagnostics and predictive maintenance of the resources). We present RSCDF as temporal and contextual extensions of RDF and discuss a State-Symptom-DiagnosisDecision-Maintenance model as the basis for RSCDF schema. Then we present RSCDF schema itself and show some examples.

Abstract: One of the main difficulties in Learning Management Systems (LMS) is how to maintain the balance between learning and management. The primary aim of LMSs was to assist educators to manage learning. As a result of this way of thinking, features such as managing content, roster students, and similar, mostly were promoted. As time goes by, the accent of research is more in student experiences and efficacy of the tools. On the other hand some educators argue that as a result of evolving Web 2.0 applications, students can be better served by an LMS alternative, a toolbox of web resources that could contain social bookmarking tools, document sharing applications, social networking applications, timeline tools, and media options offered in the cloud. Primary to this approach is the idea that students should become more experienced with today's technology tools since these skills will be practical in the workplace. As a result, some institutions have begun to offer LMS alternatives, and some teachers are using these options to support their students' learning. The paper analyses both sides of the latest approach toward LMSs and presents their implications for teaching and learning.

Abstract: 1 IntroductionThe most popular and basic techniques for accessing the Web are browse and searchFaceted search, also called faceted navigation or faceted browse, is a third technique thatenables users to access Web pages from different facets A faceted search system can displaydifferent facets of the same set of Web pages at different steps in a search Faceted search isalso a way to deal with the size of Web contents since it can locate the sought for contentthrough multi-step refinements on classifications from multiple facets [2, 5, 9] Facetedsearch has been adopted in many applications such as e-learning, e-business, online library,and general search engines However, research on its fundamental theory is neglected Manyimportant problems are still open, such as the underlying theory for organizing resourcesfrom multiple dimensions, the generation of appropriate interactive query process, andadaptive interfaces Research on the fundamental model and associated theory is criticalfor the development of faceted searchClassification and link are the most basic intelligent mechanisms for managing objectsand structuring space Making use of classification and link is the natural way to manageWeb resources Multi-dimensional classifications, called Resource Space Model, have beenused to manage various resources [13–16] It is the theory, model and method that canorganize Web resources for faceted navigationClassifications form and evolve with the evolution of society Humans use words toindicate classifications to realize effective interaction [12] Manually classifying Web re-sources is a straightforward approach to organize resources for faceted navigation Togenerate facets automatically is the key to efficient faceted navigation In Web navigationapplications, to extract the appropriate words that indicate classifications is the key togenerating facets for automatically faceted Web navigation

Abstract: Users’ freedom resulted into the exponential growth of the World Wide Web containing billions of web pages but aggravated the problem of finding specific information precisely. Semantic Web technology is deemed as solution to the problem by annotating web resources with metadata. To realize the idea of Semantic Web, several technologies are invented in the past several years. Resource Description Framework (RDF) and Topic Maps are the two prominent technologies developed for turning the vision of Semantic Web into reality by facilitating knowledge integration and sharing. Both RDF and Topic Maps provides constructs to enrich web contents with semantic information to make them useful for both humans and machines. However, RDF and Topic Maps differs and suffers from the interoperability problem due to difference in their fundamental architectures, approaches, techniques, and owner organizations and might result into two separate islands on the Web of future. To facilitate developers in understanding the technologies and accelerate their development process to produce valuable applications, a number of tools are invented by the researchers, academia, and organizations for both RDF and Topic Maps. This paper presents a comprehensive overview of the tools by investigating, analysing, classifying, and comparing them using a set of parameters. The available tools are broadly classified into three categories: storage, editing, and visualization, and their interoperability problem is addressed. We hope that this paper will provide a compact platform for both researchers and users to help them in selecting tools rich enough to satisfy their needs effectively.

Abstract: The invention provides a Web access optimizing device and method. The Web access optimizing device is applied to a Web server. The method includes the steps that a Web resource request from a client side browser is analyzed, and whether self-adaptive resource files exist in Web resources in the request or not is judged; if the self-adaptive resource files exist in Web resources in the request, whether the time of the Web resource request of the client side browser is in a cache update time window of the self-adaptive resource files or not is judged; if the time of the Web resource request is in a cache update time window, paths of the self-adaptive resource files are changed, and the changed paths of the self-adaptive resource files are added to the content of the Web resources and fed back to the client side browser. By the adoption of the Web access optimizing device and method, usability of Web applications and Web access experience of users are improved.

Abstract: We built a researcher identifier management system called the Researcher Name Resolver (RNR) to assist with the name disambiguation of authors in digital libraries on the Web. RNR, which is designed to cover all researchers in Japan, is a Web-oriented service that can be openly connected with external scholastic systems. We expect it to be widely used for enriched scholarly communications. In this paper, we first outline the conceptual framework of RNR, which is jointly focused on researcher identifier management and Web resource linking. We based our researcher identifier scheme on the reuse of multiple sets of existing researcher identifiers belonging to the Japanese grant database KAKEN and the researcher directory ReaD & Researchmap. Researcher identifiers are associated by direct links to related resources on the Web through a combination of methods, including descriptive mapping, focused crawling on campus directories and researcher identification by matching names and affiliations. Second, we discuss our implementation of RNR based on this framework. Researcher identifiers construct uniform resource identifiers to show Web pages that describe researcher profiles and provide links to related external resources. We have adapted Web-friendly technologies--e.g., OpenSearch and the RDFs of Linked Data technology--in this implementation to provide Web-friendly services. Third, we discuss our application of RNR to a name disambiguation task for the search portal of the Japanese Institutional Repositories Online to determine how well the researcher identifier management system cooperates with external systems. Finally, we discuss lessons learned from the entire project as well as the future development directions we intend to take.

Abstract: The eScholarship Research Centre (ESRC) has produced EAC-CPF XML outputs from web resources for many years. Using the Find & Connect web resource project as a case study, this article explores the uses that the ESRC has developed for EAC-CPF data, including online presentation of descriptive contextual information, as the basis of online search services, as a system-independent and preservable copy of research datasets, and as a harvestable data source for interoperability with other datasets. Also covered are expanded uses of the EAC-CPF schema for describing a wider range of contextual entities beyond corporate bodies, persons and families.

Abstract: Briefly, embodiments of methods and/or systems of providing relevant and diverse recommendations are disclosed. For one embodiment, as an example, a system may extract structured and/or semi-structured parameters from web resources obtained from interaction logs comprising records of browsing sessions. Content from extracted parameters may be compared, using an ontology, to find relationships among web resources and query resources.

Abstract: Continuing education and training are essential for a vital and productive organization and for employee adaptability and job satisfaction. Libraries of all types are organizations that value learning. Training and development for library employees is expensive, and can be out of reach for smaller institutions, or libraries that are not well-funded. Regardless of funding, libraries of all types can benefit from the wide variety of training and continuing education opportunities available on the Web as open access resources. These include documentation, journal articles, reports and white papers, online courses, videos, podcasts, and so on, from many different reliable sources. This chapter surveys major sources and types of open access online learning, and evaluates their usefulness for library employees.

Abstract: Management practices for paper map collections have evolved with changes in the modern library climate. In the past, management of “non-special” map collections were focused on efficiently processing and storing large influxes of materials. Now that fewer maps are published in paper format and more are available online, collection practices should reflect the increasingly special status of paper maps. The author examined numerous resources which together build an understanding of best practices for the physical care and management of a paper rare map collection in the 21st century. Resources consulted include publications, web resources, interviews with practicing librarians, videos, and observations made at numerous facilities. The purpose of the article is to guide collection managers in evaluating their own practices and facilities and identifying potential changes.

Abstract: A method and system to facilitate a secure user interaction with a web resource on a primary device by establishing a connection on a communication channel between the primary device and the web resource. The user interacts with a secondary device utilizing a separate communication channel to perform the action.

Abstract: Well-designed formative feedback can enhance both instructor teaching and student learning with "two-way" feedback. The project goal was to use web-enabled tools and resources that facilitate the strategies, practices, and assessments that use frequent formative feedback to improve student attitude, learning, and achievement. The research question is, "What is the effect of formative feedback, engagement pedagogy, and web-enabled resources on student outcomes across diverse settings?" The web tools included: a two-way student-reflection and instructor-feedback web data collection tool; a web vocabulary building tool with e-flash cards and e-vocabulary games; and a slide sharing site that has the slide sets from each tutorial video. Assessment was done for student: resource use with Student Resource Value Survey (SRVS); attitude and impact with a Student Impact Value Survey (SIVS); achievement with a Materials Concept Inventory (MCI); and persistence with a week-2 to week-15 head count. Results showed: large changes in certain resources from SRVS; very positive values of motivation and utility from SIVS; a 36% moderate Hake gain from the MCI; and 97% persistence. Overall, it has been demonstrated that the principles and tools of a web and engagement curricular enhancement model can have good outcomes with the potential for implementation in other disciplines.

Abstract: University of Minnesota Ph.D. dissertation. August 2014. Major; Computer science. Advisor: Jaideep Srivastava. 1 computer file (PDF); xi, 144 pages.

Abstract: The multi-disciplinary nature of Web Science and the large size and diversity of data collected and studied by its practitioners has inspired a new type of Web resource known as the Web Observatory. Web observatories are platforms that enable researchers to collect, analyze and share data about the Web and to share tools for Web research. At the Boston Web Observatory Workshop 2013, a semantic model for describing Web Observatories was drafted and an extension to the schema.org microdata vocabulary collection was proposed. This paper details our implementation of the proposed extension, and how we have applied it to the Web Observatory Portal created by the Tetherless World Constellation at Rensselaer Polytechnic Institute (TWC RPI). We recognize this effort to be the "first-step" in the construction, evaluation and validation of the Web observatory model and not the final recommendation. Our hope is that this extension recommendation and our initial implementation sparks additional discussion among the Web Science community of on whether such direction enables Web Observatory curators to better expose and explain their individual Web Observatories to others, thereby enabling better collaboration between researchers across the Web Science community.

Abstract: The invention provides a device and method for Web access optimization. The method includes the steps that an initial page Web resource request is received from a client side browser; a resource necessary to be used for loading an initial page serves as a first response resource, and the first response resource is carried in a Web response to be sent to the client side browser; after a global page Web resource request is received from the client side browser, a preset second response resource is carried in the Web response to be sent to the client side browser and includes a resource necessary for loafing a global page. According to the technical scheme, the initial page request of the client side browser is low in load and high in loading speed, and then user experience is improved.

Abstract: The time spent using a web browser on a wide variety of tasks such as research activities, shopping or planning holidays is relevant. Web pages visited by users contain important hints about their interests, but empirical evaluations show that almost 40-50% of the elements of the web pages can be considered irrelevant w.r.t. the user interests driving the browsing activity. Moreover, pages might cover several different topics. For these reasons they are often ignored in personalized approaches. We propose a novel approach for selectively collecting text information based on any implicit signal that naturally exists through web browsing interactions. Our approach consists of three steps: (1) definition of a DOM-based representation of visited pages, (2) clustering of pages according with a tree edit distance measure and (3) exploiting the acquired evidence about the user behaviour to better filtering out irrelevant information and identify relevant text related to the current needs. A comparative evaluation shows the effectiveness of the proposed approach in retrieving additional web resources related to what the user is currently browsing