The exponential growth in medical imaging volumes necessitates scalable, reliable diagnostic support systems capable of augmenting clinical workflows. This paper presents a systematic quantitative evaluation of state-of-the-art Multimodal Large Language Models (MLLMs) for radiology Visual Question Answering (VQA), a task requiring integrated visual perception and clinical reasoning. We benchmark five leading models-GPT4o, Gemini 2.5 Pro, Qwen 2.5-VL (72B), LLaVA 13B, and Llama 3.2 Vision (90B)-on the VQA-RAD dataset under a rigorous zero-shot protocol with standardized prompts and comprehensive precision-recall-F1 evaluation. Our empirical analysis reveals that Gemini 2.5 Pro achieves superior balanced performance (F1$=0.77$, Accuracy$=0.76$, Recall$=0.81$), while Qwen 2.5-VL attains the highest precision (0.87) at the cost of reduced recall, exhibiting conservative prediction characteristics. These outcomes demonstrate that general-purpose MLLMs can perform competitively with specialized medical models in tasks such as modality and organ recognition but still struggle with abnormality detection and complex clinical reasoning. The findings reinforce that MLLMs currently serve best as assistive decision-support tools rather than autonomous diagnostic agents and highlight the potential of retrieval-augmented and context-aware strategies for improving clinical reliability and interpretability.

Automated GUI testing can play a crucial role in uncovering faults within web applications. In this context, scriptless testing can streamline the process by automatically generating and executing test cases that explore GUI elements. However, discovering non-repetitive actionable elements within complex web pages remains a challenging task. Often, testers need to employ manual strategies to facilitate element discovery and avoid redundancies. In this paper, we introduce a novel automatic approach, Unique Actionable Element Search (UAES), designed to provide a way to semantically differentiate elements in web applications. By leveraging common HTML properties and semantically expressive locators, UAES streamlines the process of discovering and distinguishing actionable elements, enhancing the efficiency and accuracy of systematic GUI testing. We conducted two empirical studies involving four open-source and twenty industrial web applications. They compared the results of our approach to those obtained using explicit markups included by experienced testers. The results indicate that our approach discovered 94.25% of the elements manually identified by the testers, while also uncovering a significant number of elements (46.86% of all elements discovered) that the testers overlooked.

In the last years, many research works have developed Data Warehouses to perform trajectory data analysis. An essential problem in this context is how to represent trajectory data in the data warehouse. Another critical challenge consists of enhancing trajectory data by describing the semantics of each movement, enabling these systems to solve queries concerning multiple aspects of the trajectories. This paper presents a trajectory framework that encompasses the main steps in the transformation process, semantic enrichment, and data analysis. In addition to semantic information already commonly dealt with by other research in the area, our framework also includes other semantic information based on a context taxonomy present in the literature. The raw trajectories are semantically enriched, summarized in a multidimensional model, and made available for analysis by an exploratory SOLAP tool. Furthermore, we explore new information on the web to enhance the analytics process.

Traditionally, OLAP tools have been performed on static and structured data sources. Nowadays, there is plenty of data in different formats, such as the Resource Description Framework (RDF), a semi-structured and semantically-rich format external to the BI infrastructure. Although such data formats are enriched by semantics, performing data analysis is challenging. As a result, the Exploratory OLAP field has emerged for discovery, acquisition, integration and querying such data, aiming at performing a complete and effective analysis on both internal and external data. In this paper, we propose an innovative Exploratory OLAP approach that integrates on the fly semantic semi-structured and structured data sources.

A geographic search engine should be capable of identifying, indexing and retrieving the geographic context present in a considerable portion of the Web content, which is not expected in classic search engines. This type of search engine is characterised by a set of features, such as toponym recognition and resolution, geographic scope modelling, and relevance ranking. Ranking the results returned by a user query is a crucial task for this kind of system. In this paper, we present a method of ranking Web documents, which consider both the documents' geographic context and information collected from microblogs. This method also comprises an algorithm to estimate the affinity between locations and social network users, based on their interactions. The proposed method was evaluated within the news domain, using a system prototype which recognises geographic information contained in the news content and integrates information collected from Twitter.

There are several research works using web semantic to improve integration and retrieval of geospatial data provided by spatial data infrastructures (SDI) and geoportals. However, an important open issue that deserves more research efforts is how to implement the process of semantic annotation, in which geospatial data are associated to concepts formally defined in application ontologies. Since the volume of the data provided by a SDI is usually very large, it is necessary to provide mechanisms that are able to perform semantic annotation automatically (or semi-automatically). Aiming to solve this limitation, this article proposes a method that uses the open data provided by DBpedia, as well ideas proposed in the classic information retrieval, to provide automatic semantic annotation of feature types provided by SDIs. We used geospatial data gathered from the catalog service of a real SDI in order to evaluate our proposed method. The results obtained from initial experiments showed that our approach is viable, since it can generate semantic annotations to many feature types with an acceptable precision rate.

Price survey services assist consumers interested in purchasing a new product online. Nevertheless, even being efficient in consulting the suppliers and returning up-to-date information about price and availability of products, current services still provide trivial and incomplete information concerning the price dispersion and the price for the final consumer. Usually, only instant prices are provided. These prices change rapidly and do not include shipping costs, which are usually paid by the consumers. In this context, this work proposes an approach based on Big Data to introduce the spatial context supplier-consumer in comparative price surveys and also to store and present the price variation history of the products as a function of their spatial and temporal features. A case study is carried out in order to validate the proposal and to highlight the advantages of this approach for the users of such services.

Recently, digital photography is definitely becoming very popular. Thus, there is a large increase in the number of captured photos, which demands better methods to manage this data. An approach is to allow the user to annotate some information in those photos, for instance, the people who appear in them. In this paper, two solutions are proposed in order to assist the user in the process of annotating people in photos, using mobile devices. The first one, called on-line, accesses a server to produce the results and the other one, called off-line, utilizes the mobile device itself to compute the results. Finally, a comparative study between the solutions, using parameters such as hit rate and performance, is presented.

Nowadays the photo-capturing devices are no longer limited to digital cameras but include mobile phones, PDAs and others. This is leading to a new problem: a very large number of digital photos captured and chaotically stored in multiple locations without being annotated. This paper presents a new system, called PhotoGeo, for self-organization of georeferenced photos. The proposed system uses metadata and external sources to organize the photos. These external sources include a map-based social network and user’s calendar. Furthermore, spatial clustering and temporal segmentation techniques are used to join the photos into clusters with similar features.

In this paper we present a system for the creation of both tourist and business trip plans. SEI-Tur is a recommender system that uses Web services in order to create complete and consistent tours, based on user preferences, combining travel, lodging, food supply and local event programming. For each component of the tour, a domain ontology has been developed and the discovery and selection of the best services is based on a sophisticated service selection and composition tool.

Advances in mobile devices, sensors and wireless networks have motivated the development of context-aware applications. Such mobile applications use sensors for monitoring several features such as: location, temperature, velocity, weather, traffic jam, noise, air pollution, and so on. This monitoring enables service provision according to a given context. These applications are known as context-aware systems. Context-aware applications are sensible to user necessities, personalized according to his profile, requirements and context. They evaluate user environment and push information that is relevant to user context. In this paper we present Omnipresent, which is a service-oriented architecture for context-aware applications. Omnipresent may be accessed from either mobile devices or Web browsers, and it is based on Web services and well-established standards for LBS applications such as those proposed by the OpenGeoSpatial Consortium. Omnipresent offers several services: map presentation, routing, advertisement, and also works as a reminder tool.

Visual Query Systems (VQS) are becoming a very attractive field of research, especially for advanced database systems such as spatial and temporal ones. However, most of the visual query proposals support either spatial or temporal data but not both. This paper presents a new VQS which supports querying spatio-temporal data. The main idea is to provide a web-based, user-friendly, and visual environment for querying generic spatio-temporal databases. Therefore end users do not need to worry about neither data schemas nor query language syntax. The proposed VQS is based on well-established standards such as SQL and OpenGIS, and it is flexible to be used in many database systems with support for spatial data.

Decision support systems (DSS) may be enhanced qualitatively if they are able to also deal with spatial dimensions and measures. Regardless the evident importance of using data warehousing and OLAP in DSS, the incorporation of spatial dimensions and measures enables to locate more efficiently tendencies in a given application domain, by using dynamic maps with zooming, panning, aggregation and other spatial functionalities. Therefore, it is necessary to converge two relatively consolidated technologies: Data Warehousing and Geographical Information Systems. This integration gives raise to a new research area called Spatial Data Warehousing (SDW), which introduces new research challenges. This paper proposes a novel logical multidimensional model suitable for SDW, which is implemented on the top of an object-relational database system with support for spatial data. Moreover, the paper addresses query optimization techniques to enhance performance and it describes a prototype, which has been built to validate the proposed ideas.

The use of mobile applications is increasing rapidly. Particularly, mobile applications for decision support systems, such as mobile data warehouses, are very attractive. However, mobile application imposes new requirements such as disconnection from the data sources. These requirements force the use of incremental update algorithms in order to update the mobile hosts. We extend the idea of incremental update algorithms with an optimized hierarchy of views which enables improving performance on mobile data warehouses. Moreover, we prove that other aggregate functions apart from the standard ones, in particular, standard deviation, variance, and ratiotoreport, are suitable for being used in incremental update algorithms.