The proliferation of scientific literature presents an increasingly significant challenge for researchers. While Large Language Models (LLMs) offer promise, existing tools often provide verbose summaries that risk replacing, rather than assisting, the reading of the source material. This paper introduces InsightGUIDE, a novel AI-powered tool designed to function as a reading assistant, not a replacement. Our system provides concise, structured insights that act as a “map” to a paper's key elements by embedding an expert's reading methodology directly into its core AI logic. We present the system's architecture, its prompt-driven methodology, and a qualitative case study comparing its output to a general-purpose LLM. The results demonstrate that InsightGUIDE produces more structured and actionable guidance, serving as a more effective tool for the modern researcher.

Heterogeneous information networks (HINs) represent different types of entities and relationships between them. Exploring and mining HINs relies on metapath queries that identify pairs of entities connected by relationships of diverse semantics. While the real-time evaluation of metapath query workloads on large, web-scale HINs is highly demanding in computational cost, current approaches do not exploit interrelationships among the queries. In this paper, we present Atrapos, a new approach for the real-time evaluation of metapath query workloads that leverages a combination of efficient sparse matrix multiplication and intermediate result caching. Atrapos selects intermediate results to cache and reuse by detecting frequent sub-metapaths among workload queries in real time, using a tailor-made data structure, the Overlap Tree, and an associated caching policy. Our experimental study on real data shows that Atrapos accelerates exploratory data analysis and mining on HINs, outperforming off-the-shelf caching approaches and state-of-the-art research prototypes in all examined scenarios.

Graphs are ubiquitous and ever-present data structures that have a wide range of applications involving social networks, knowledge bases and biological interactions. The evolution of a graph in such scenarios can yield important insights about the nature and ac- tivities of the underlying network, which can then be utilized for applications such as news dissemination, network monitoring, and content curation. Capturing the continuous evolution of a graph can be achieved by long-standing sub-graph queries. Although, for many applications this can only be achieved by a set of quer- ies, state-of-the-art approaches focus on a single query scenario. In this paper, we therefore introduce the notion of continuous multi-query processing over graph streams and discuss its appli- cation to a number of use cases. To this end, we designed and developed a novel algorithmic solution for efficient multi-query evaluation against a stream of graph updates and experimentally demonstrated its applicability. Our results against two baseline approaches using real-world, as well as synthetic datasets, confirm a two orders of magnitude improvement of the proposed solution.

Due to the rapidly increasing number of scientific articles, finding valuable work for further research has become tedious and time consuming. To alleviate this issue, search engines have used citation-based article impact ranking. However, most engines rely on very simplistic impact measures (usually the citation count) and make the problematic assumption that there is a one-size-fits-all impact measure. To address these problems, we present BIP! Finder, a search engine that facilitates the identification of valuable articles by exploiting two different impact measures, each capturing a different aspect of the article impact. In addition, BIP! Finder provides many useful features (article comparison, intuitive visualisations, article bookmarking mechanism, etc.) making it a powerful addition to the researcher's toolbox.

The clear, social, and dark web have lately been identified as rich sources of valuable cyber-security information that -given the appropriate tools and methods-may be identified, crawled and subsequently leveraged to actionable cyber-threat intelligence. In this work, we focus on the information gathering task, and present a novel crawling architecture for transparently harvesting data from security websites in the clear web, security forums in the social web, and hacker forums/marketplaces in the dark web. The proposed architecture adopts a two-phase approach to data harvesting. Initially a machine learning-based crawler is used to direct the harvesting towards websites of interest, while in the second phase state-of-the-art statistical language modelling techniques are used to represent the harvested information in a latent low-dimensional feature space and rank it based on its potential relevance to the task at hand. The proposed architecture is realised using exclusively open-source tools, and a preliminary evaluation with crowdsourced results demonstrates its effectiveness.

In the information filtering paradigm, clients subscribe to a server with continuous queries that express their information needs and get notified every time appropriate information is published. To perform this task in an efficient way, servers employ indexing schemes that support fast matches of the incoming information with the query database. Such indexing schemes involve (i) main-memory trie-based data structures that cluster similar queries by capturing common elements between them and (ii) efficient filtering mechanisms that exploit this clustering to achieve high throughput and low filtering times. However, state-of-the-art indexing schemes are sensitive to the query insertion order and cannot adopt to an evolving query workload, degrading the filtering performance over time. In this paper, we present an adaptive trie-based algorithm that outperforms current methods by relying on query statistics to reorganise the query database. Contrary to previous approaches, we show that the nature of the constructed tries, rather than their compactness, is the determining factor for efficient filtering performance. Our algorithm does not depend on the order of insertion of queries in the database, manages to cluster queries even when clustering possibilities are limited, and achieves more than 96 percent filtering time improvement over its state-of-the-art competitors. Finally, we demonstrate that our solution is easily extensible to multi-core machines.

Privacy of Internet users is at stake because they expose personal information in posts created in online communities, in search queries, and other activities. An adversary that monitors a community may identify the users with the most sensitive properties and utilize this knowledge against them (e.g., by adjusting the pricing of goods or targeting ads of sensitive nature). Existing privacy models for structured data are inadequate to capture privacy risks from user posts.

Among the basic research tools for (bio)medical science are epidemiological studies that typically involve a number of hospitals, clinics, and research centres scattered around the world, and are often referred to as multi-centre studies. Clearly, the effectiveness and importance of a multi-centre study increases with the number of participating centres and enrolled patients, but at the same time this natural distribution in the production of research data requires sophisticated data/knowledge management infrastructures to support the participating units. This kind of infrastructure is not only expensive to build and maintain, but also cannot be reused as it is often tailored to a specific study. In this work, we present a cloud-based system, that allows users without any computer science background to design, deploy, and administer platforms aimed for managing, sharing, and analysing clinical data from multi-centre studies. The proposed system provides a zero-administration, zero-cost online data/knowledge management tool that (i) enhances re-usability by introducing study templates, (ii) supports (bio)medical needs through specialised data types able to capture specialised knowledge like repeated therapies or treatments, and (iii) emphasises data filtering/export through an expressive yet simple graphical query engine.

Publishing data about individuals, in a privacy-preserving way, has led to a large body of research. Meanwhile, algo-rithms for anonymizing datasets, with relational or trans-action attributes, that preserve data truthfulness, have at-tracted significant interest from organizations. However, se-lecting the most appropriate algorithm is still far from triv-ial, and tools that assist data publishers in this task are needed. In response, we develop SECRETA, a system for analyzing the effectiveness and efficiency of anonymization algorithms. Our system allows data publishers to evalu-ate a specific algorithm, compare multiple algorithms, and combine algorithms for anonymizing datasets with both re-lational and transaction attributes. The analysis of the algo-rithm(s) is performed, in an interactive and progressive way, and results, including attribute statistics and various data utility indicators, are summarized and presented graphically. 1.

Users nowadays need to manage large amounts of information, including documents, e-mails, contacts, and multimedia content. To facilitate the tasks of organisation, maintenance, and retrieval of personal information, a number of semantics-based methods have emerged; these methods employ (personal) ontologies as an underlying infrastructure for organising and querying the personal information space. In this paper we present OntoFM, a novel personal information management tool that offers a mindmap-inspired interface to facilitate user interactions with the information base. Besides serving as an information retrieval aid, OntoFM allows the user to specify and update the semantic links between information items, constituting thus a complete personal information management tool. Keywords personal information management, mindmaps, file manager,

Among the basic research tools for (bio)medical science are epidemiological studies that typically involve a number of hospitals, clinics, and research centres scattered around the world, and are often referred to as multi-centre studies. Clearly, the effectiveness and importance of a multi-centre study increases with the number of participating centres and enrolled patients, but at the same time this natural distribution in the production of research data requires sophisticated data management infrastructures to support the participating units. This kind of infrastructure is not only expensive to build and maintain, but also cannot be reused as it is often tailored to a specific study. In this work, we present a cloud-based system, coined CLOUDSTUDY, that allows users without any computer science background to design, deploy, and administer platforms aimed for managing, sharing, and analysing clinical data from multi-centre studies. The CLOUDSTUDY system provides a zero-administration, zero-cost online tool for creating multi-centre studies that (i) enhances re-usability by introducing study templates, (ii) supports (bio)medical needs through specialised data types, and (iii) emphasises data filtering/export through an expressive yet simple graphical query engine.

Semantic overlay networks cluster peers that are semantically, thematically or socially close into groups, by means of a rewiring procedure that is periodically executed by each peer. This procedure establishes new connections to similar peers and disregards connections to peers that are dissimilar. Retrieval effectiveness is then improved by exploiting this information at query time (as queries may address clusters of similar peers). Although all systems based on semantic overlay networks apply some rewiring technique, there is no comprehensive study showing the effect of rewiring on system’s performance. In this work, a framework for studying the attribution of rewiring strategies in semantic overlay networks is proposed. A generic approach to rewiring is presented and several variants of this approach are reviewed and evaluated. We show how peer organisation is affected by the different design choices of the rewiring mechanism and how these choices affect the performance of the system overall (both in terms of communication overhead and retrieval effectiveness). Our experimental evaluation with real-word data and queries confirms the dependence between rewiring strategies and retrieval performance, and gives insights on the trade-offs involved in the selection of a rewiring strategy.

In the information filtering paradigm, clients subscribe to a server with continuous queries or profiles that express their information needs. Clients can also publish documents to servers. Whenever a document is published, the continuous queries satisfying this document are found and notifications are sent to appropriate clients. This article deals with the filtering problem that needs to be solved efficiently by each server: Given a database of continuous queries db and a document d , find all queries q ∈ db that match d . We present data structures and indexing algorithms that enable us to solve the filtering problem efficiently for large databases of queries expressed in the model AWP . AWP is based on named attributes with values of type text, and its query language includes Boolean and word proximity operators.

In this demonstration paper we present MAPS, a novel system that combines approximate information retrieval and filtering functionality in a peer-to-peer setting. In MAPS, a user is able to submit one-time and continuous queries, and receive matching resources and notifications from selected information sources. The selection of these sources in the retrieval case is based on well-known resource selection techniques for peer-to-peer query routing, while in the filtering case a combination of resource selection and novel behavior prediction techniques using time-series analysis of publisher statistics is used. The integration of the two functionalities is done in a seamless way utilizing the same machinery: a conceptually global, but physically distributed directory of statistics about information sources based on distributed hash tables.

Semantic overlay networks are an instance of unstructured overlays, where peers that are semantically, thematically, or socially close are organized into groups to exploit similarities at query time. In this work we present Clouds, a novel P2P search infrastructure for providing anonymous and censorship resistant search functionality in such networks. Although we utilize semantic overlays to exploit their retrieval capabilities, our framework is general and can be applied to any unstructured overlay.

Information filtering, also referred to as publish/subscribe, complements one-time searching since users are able to subscribe to information sources and be notified whenever new documents of interest are published. In approximate information filtering only selected information sources, that are likely to publish documents relevant to the user interests in the future, are monitored. To achieve this functionality, a subscriber exploits statistical metadata to identify promising publishers and index its continuous query only in those publishers. The statistics are maintained in a directory, usually on a per-keyword basis, thus disregarding possible correlations among keywords. Using this coarse information, poor publisher selection may lead to poor filtering performance and thus loss of interesting documents.1

We study the problem of continuous relational query processing in Internet-scale overlay networks realized by distributed hash tables. We concentrate on the case of continuous two-way equi-join queries. Joins are hard to evaluate in a distributed continuous query environment because data from more than one relations is needed, and this data is inserted in the network asynchronously. Each time a new tuple is inserted, the network nodes have to cooperate to check if this tuple can contribute to the satisfaction of a query when combined with previously inserted tuples. We propose a series of algorithms that initially index queries at network nodes using hashing. Then, they exploit the values of join attributes in incoming tuples to rewrite the given queries into simpler ones, and reindex them in the network where they might be satisfied by existing or future tuples. We present a detailed experimental evaluation in a simulated environment and we show that our algorithms are scalable, balance the storage and query processing load and keep the network traffic low.

We study the complexity of query satisfiability and entailment for the Boolean information retrieval models WP and AWV using techniques from propositional logic and computational complexity. WP and AWV can be used to represent and query textual information under the Boolean model using the concept of attribute with values of type text, the concept of word, and word proximity constraints. Variations of WP and AWP are in use in most deployed digital libraries using the Boolean model, text extenders for relational database systems (e.g., Oracle 10g), search engines, and P2P systems for information retrieval and filtering

We study the problem of offering publish/subscribe functionality on top of structured overlay networks using data models and languages from IR. We show how to achieve this by extending the distributed hash table Chord and present a detailed experimental evaluation of our proposals.

In the selective dissemination of information (or publish/subscribe) paradigm, clients subscribe to a server with continuous queries (or profiles) that express their information needs. Clients can also publish documents to servers. Whenever a document is published, the continuous queries satisfying this document are found and notifications are sent to appropriate clients. This paper deals with the filtering problem that needs to be solved effciently by each server: Given a database of continuous queries db and a document d, find all queries q ∈ db that match d. We present data structures and indexing algorithms that enable us to solve the filtering problem efficiently for large databases of queries expressed in the model AWP which is based on named attributes with values of type text, and word proximity operators.

No abstract available.