Cláudio de Souza Baptista

Towards a logical multidimensional model for spatial data warehousing and OLAP

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.

Improving geographic information retrieval in spatial data infrastructures

In recent years, spatial data infrastructures (SDIs) have gained great popularity as a solution to facilitate interoperable access to geospatial data offered by different agencies. In order to enhance the data retrieval process, current infrastructures usually offer a catalog service. Nevertheless, such catalog services still have important limitations that make it difficult for users to find the geospatial data that they are interested in. Some current catalog drawbacks include the use of a single record to describe all the feature types offered by a service, the lack of formal means to describe the semantics of the underlying data, and the lack of an effective ranking metric to organize the results retrieved from a query. Aiming to overcome these limitations, this article proposes SESDI (Semantically-Enabled Spatial Data Infrastructures), which is framework that reuses techniques of classic information retrieval to improve geographic data retrieval in a SDI. Moreover, the framework proposes several ranking metrics to solve spatial, semantic, temporal and multidimensional queries.

PhotoGeo: a photo digital library with spatial-temporal support and self-annotation

The recent popularity of digital cameras has posed a new problem: how to efficiently store and retrieve the very large number of digital photos captured and chaotically stored in multiple locations without any annotation. This paper proposes an infrastructure, called PhotoGeo, which aims at helping users with the people photo annotation, event photo annotation, storage and retrieval of personal digital photos. To achieve the desired objective, PhotoGeo uses new algorithms that make it possible to annotate photos with the key metadata to facilitate their retrieval, such as: the people who were shown in the photo (who); where it was captured (where); the date and time of capture (when); and the event that was captured. The paper concludes with a detailed evaluation of these algorithms.

Querying spatio-temporal databases using a visual environment

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.

Compass clustering: a new clustering method for detection of points of interest using personal collections of georeferenced and oriented photographs

Knowledge discovery in large online photographic repositories has been an active area of research in recent years. This is due to the great popularization of devices equipped with image capture, such as digital cameras, smartphones and tablets. Moreover, the image files generated by those devices are easily spread out on the Web through social networking sites. Typically, the photos stored in these repositories bear valuable metadata, such as, geographic coordinates, timestamp, and camera orientation. This information can be used for many interesting data mining tasks, such as detection of points-of-interest (POIs) and trip planning. This paper introduces Compass Clustering, a new clustering algorithm for detecting POIs in georeferenced and oriented photo repositories. Most of the state-of-the-art approaches for POI detection cluster photos based solely on their geographic proximity. However, in many cases, the POIs are within a certain distance from the point where the photo was taken, that is, not in the exact camera location but in the direction it is pointing to, and thus many photos would be erroneously classified by existing methods. Therefore, we propose to exploit the camera orientation in order to identify more reliable POIs that reflect the real intention of people when taking photos. We evaluated our approach on a collection of more than 8,000 georeferenced and oriented photos collected from Flickr.

A Comparison of SVM Versus Naive-Bayes Techniques for Sentiment Analysis in Tweets: A Case Study with the 2013 FIFA Confederations Cup

The widespread of social communication media on the Web has made available a large volume of opinionated textual data stored in digital format. These media constitute a rich source for sentiment analysis and understanding of the opinions spontaneously expressed. Traditional techniques for sentiment analysis are based on POS Tagger. Considering the Portuguese language, the use of POS Tagging ends up being too costly, due to the complex grammatical structure of this language. Faced with this problem, a case study is carried out in order to compare two techniques for sentiment analysis: a SVM versus Naive-Bayes classifiers. Our study focused on tweets written in Portuguese during the 2013 FIFA Confederations Cup, although our technique could be applied to any other language. The achieved results indicated that the SVM technique surpassed the Naive-Bayes one, concerning performance issues.

Using Federated Catalogs to Improve Semantic Integration among Spatial Data Infrastructures

Nowadays, Spatial Data Infrastructures (SDIs) play an important role in government agencies, at different levels: global, national, and local. They aim to improve the management and sharing of geospatial data. Nonetheless, these SDIs have been developed as information islands, in which a users query is compared to metadata described only in their own catalog services. The lack of interaction among SDIs limits the potential of these infrastructures in providing geospatial data to a larger audience. This article presents a distributed architecture, based on a federation of SDIs which interact among themselves, using query propagation. This propagation facilitates data discovery and sharing. We also describe a distributed query processing service used to enable the resource discovery in distributed infrastructures.

A Model for Geographic Knowledge Extraction on Web Documents

There is an increasing interest on doing research in the field of information retrieval which aims to incorporate new dimensions, apart from text based retrieval, to the Web search engines. Geographical Information Retrieval (GIR) aims to index Web resources using a geographic context. The process of identifying the geographic context starts with the detection of different types of geographic references associated to the documents, as for example, the occurrence of place names. This paper presents a model for detecting geographic references in Web documents based on a set of heuristics. Moreover, new concepts and methods for disambiguation of many places with the same name are addressed. Finally, a prototype was built, called GeoSEn which aimed to validate the effectiveness of the proposed model.

Geographic scope modeling for web documents

Geographic Information Retrieval (GIR) has become a very attractive area of research. GIR is a specialization of a traditional information retrieval system, which may index and search Web documents based on their spatial footprints. Research in this new field may be categorized into crawling spatial-related documents, modeling the geographic scope of a document, indexing these documents using textual and spatial features, and the building of spatially-enabled searching and ranking. This paper presents a method for modeling the geographic scope. The proposed model is based on both the statistics collected from the detected references; and the spatial distribution of the places involved in a given document. This model aims to simplify the indexing and searching processes. Furthermore, the number of spatial operations is reduced, as a consequence the overall performance is improved.

Using Open Source GIS in e-Government Applications

Geographical Information System (GIS) plays an important role in e-government applications once that all government information has a geographic dimension – a street address, a transportation road, a river, a city, a state and so on. GIS apart from contributing for data integration from different data sources, enables data visualization using maps, which enhances the system usability and can help the decision-making process. This paper describes a web GIS framework that provides fast and easy deployment of e-government applications. The framework is validated by the realization of a substantial application of a Brazilian state, which contains data about health, education, industry, human development index and other socio-economic data.