Papers by Francesco Gargiulo
Ecology and the Environment, Sep 23, 2014
In the Campania region in south-west Italy, there is growing evidence, including from a World Hea... more In the Campania region in south-west Italy, there is growing evidence, including from a World Health Organization (WHO) study, that the accumulation of waste, illegal and legal, urban and industrial, has contaminated soil, water, and the air with a range of toxic pollutants including dioxins. An effective environmental monitoring system represents an important tool for the early detection of environmental violations. The IDES (Intelligent Data Extraction System) Project is a geo-environmental intelligence system (GIS), developed by the CIRA (Italian Aerospace Research Centre) with the contribution of universities and other government bodies, which aims to implement an advanced software and hardware platform for image, data and document analysis in order to support law enforcement investigations. The main IDES modules are: imagery analysis module to monitor land-use and anthropogenic changes; environmental GIS module to fuse geographical and administrative information; epidemiological domain module; and semantic search module to discover information in public sources such as blogs, social network, forums and newspapers. This paper focuses on the semantic search module and aims to provide the greatest support to those extracting possible environmental crimes, through the collection and analysis of documents from online public sources. People rarely denounce criminal activity to the authorities. On the other hand, every day sees many people exposing the status of land degradation through blogs, forums and social networks. In addition, given the public interest, journalists often document critical environmental issues. All this unstructured information is often lost, due to the difficulty in collecting and analysing it. The IDES semantic search module is an innovative solution for
EDBT/ICDT Workshops, 2014
Recommender systems help people in retrieving information that match their preferences by recomme... more Recommender systems help people in retrieving information that match their preferences by recommending products or services from a large number of candidates, and support people in making decisions in various contexts: what items to buy , which movie to watch or even who they can invite to their social network . They are especially useful in environments characterized by a vast amount of information, since they can e↵ectively select a small subset of items that appear to fit the user's needs. We present the main points related to recommender systems using multimedia data, especially for social networks applications. We also describe, as an example, a framework developed at the University of Naples "Federico II". It provides customized recommendations by originally combining intrinsic features of multimedia objects (low-level and semantic similarity), past behavior of individual users and overall behavior of the entire community of users, and eventually considering users' preferences and social interests.
International Journal of Sustainable Development and Planning, 2016
Enforcement of environmental regulation is a persistent challenge and timely detection of the vio... more Enforcement of environmental regulation is a persistent challenge and timely detection of the violations is key to holding the violators accountable. The use of remote sensing data is becoming an effective practice in the fight against environmental crimes. In this work, a novel and effective approach for the detection of potentially hazardous cattle-breeding facilities, exploiting both synthetic aperture radar and optical multispectral data together with geospatial analyses in the geographic information system (GIS) environment, is proposed. Experiments on data available for the area of Caserta (Southern Italy), show that the proposed technique provides very high detection capability, up to 90%, with a acceptable false alarm rate, becoming a useful tool in the hand of agencies engaged in the protection of territory.
2015 31st IEEE International Conference on Data Engineering Workshops, 2015
In this paper, we propose SemTree, a novel semantic index for supporting retrieval of information... more In this paper, we propose SemTree, a novel semantic index for supporting retrieval of information from huge amount of document collections, assuming that semantics of a document can be effectively expressed by a set of (subject, predicate, object) statements as in the RDF model. A distributed version of KD Tree has been then adopted for providing a scalable solution to the document indexing, leveraging the mapping of triples in a vectorial space. We investigate the feasibility of our approach in a real case study, considering the problem of finding inconsistencies in documents related to software requirements and report some preliminary experimental results.
Proceedings of 4th International Conference on Data Management Technologies and Applications, 2015
Journal of Air Transport Management, 2016
Proceedings of the 7th International Conference on Data Science, Technology and Applications, 2018
The main objective of this work is the proposal of a decentralized data structure storing a large... more The main objective of this work is the proposal of a decentralized data structure storing a large amount of data under the assumption that it is not possible or convenient to use a single workstation to host all data. The index is distributed over a computer network and the performance of the search, insert, delete operations are close to the traditional indices that use a single workstation. It is based on k-d trees and it is distributed across a network of "peers", where each one hosts a part of the tree and uses message passing for communication between peers. In particular, we propose a novel version of the k-nearest neighbour algorithm that starts the query in a randomly chosen peer and terminates the query as soon as possible. Preliminary experiments have demonstrated that in about 65% of cases it starts a query in a random peer that does not involve the peer containing the root of the tree and in the 98% of cases it terminates the query in a peer that does not contain the root of the tree. 2 RESEARCH IDEAS AND RESULTS This section introduces the problem description and our proposal to cope with it.
This work proposes a new approach to index multidimensional data based on kd-trees and proposes a... more This work proposes a new approach to index multidimensional data based on kd-trees and proposes also a novel approach to query processing. The indexing data structure is distributed across a network of "peers", where each one hosts a part of the tree and uses message passing for communication among nodes. The advantages of this kind of approach are mainly two: it is possible to i) handle a larger number of nodes and points than a single peer based architecture and ii) to run in an efficient way the elaboration of multiple queries. In particular, we propose a novel version of the k-nearest neighbor algorithm that is able to start a query in a randomly chosen peer. Furthrmore, it returns the results without traverse the peer containing the root. Preliminary experiments demonstrated that on average in about 65% of cases a query starting in a random node, does not involve the peer containing the root of the tree. Also, on average in about 98% of cases, it returns the results w...
Proceedings of 3rd International Conference on Data Management Technologies and Applications, 2014
ABSTRACT This paper describes a semantic search tool based on our experience in using a new lexic... more ABSTRACT This paper describes a semantic search tool based on our experience in using a new lexical domain ontology for aerospace integrated with an open source general purpose ontology to support aerospace engineers in the timely semantic retrieval of the knowledge. The semantic search module represents an integrated tool dedicated to the semantic search, extraction and classification of information and knowledge in aerospace domain. It describes the implementation of a disambiguation algorithm based upon these ontologies and a new interesting graphical user interface for semantic searches is presented. Furthermore, next to the domain ontology, a taxonomy for classifying aerospace documents is also proposed. The document classification algorithm that leverages the deep integration between the proposed lexical domain ontology and taxonomy is also described. Finally, some considerations about the usage of the semantic search module by the side of domain experts, semantic experts or common users are reported.
Studies in Computational Intelligence, 2015
ABSTRACT Requirements Engineering (RE) is a key discipline for the success of software projects. ... more ABSTRACT Requirements Engineering (RE) is a key discipline for the success of software projects. Consistency, completeness, and accuracy are the requirements quality properties to be guaranteed by the verification task in RE. An overview of the actual trends in RE is briefly summarized, focusing more closely on the requirements verification quality properties. Completeness results is the most difficult property to guarantee. It is hard to capture the software behavior against the whole external context. In the last years, research has focused its attention to the application of semantic Web techniques to the different tasks of RE. The adoption of ontologies seems promising to achieve the proper level of formalism and to argue on quality properties. This paper presents a survey of the main concepts that need to be accounted for requirement verification, and proposes an ontological engineering approach to demonstrate the overlapping of requirements against the external context.
2013 Eighth International Conference on P2P, Parallel, Grid, Cloud and Internet Computing, 2013
ABSTRACT The competitiveness of the aviation maintenance companies is strictly connected to the s... more ABSTRACT The competitiveness of the aviation maintenance companies is strictly connected to the security, the quality, the efficiency, the costs and the speed of the maintenance processes. This entails a critical role for the planning of maintenance activities. A specific resource allocation problem for aeronautical maintenance activities is described and a combination of a dynamic programming strategy and an agent-oriented approach for the planning of these activities is proposed. Since the dynamic programming algorithm presented is inherently parallelizable then a multi-agent strategy that allows distributed computing is proposed.it A coordination mechanism guarantees the actions of multiple agents lead to a global solution. Agent-based methods inherently overcome the limitations of centralized load-balancing and they can adaptively execute on dynamic networks of workstations.
Sustainable Development and Planning VI, 2013
IDES-Intelligent Data Extraction System-is the name of a new project, founded by Campania Regiona... more IDES-Intelligent Data Extraction System-is the name of a new project, founded by Campania Regional Government (Italy) and developed by the CIRA-Italian Aerospace Research Centre that aims at implementing an advanced tool to support Government Bodies in the discovery and localization of environmental criticalities. In detail, this project will develop a software and hardware platform for image, data and document analysis in order to support law enforcement investigations. The fight against illegal waste dumping is the first objective of the IDES project and the main goal is to develop algorithms for image analysis in order to automatically identify this environmental criticality. The first task is to identify the statistical link between specific features extracted from satellite images or acquired by airborne sensors, possibly in combination with punctual information (results of chemical analysis, physical property measurements, information pertaining to the population), and environmental criticalities (e.g. illegal dumping, illegal landfills etc). IDES will offer an integrated Geographic Information System (GIS) repository of information extracted and collected by various government bodies. Furthermore, an innovative aspect of IDES is the integration between information extracted from remotely sensed images and information extracted from textual documents through a semantic analysis. The results of this last analysis will be a taxonomydriven document searching and browsing system in which all documents are categorized and all relevant entities in documents (e.g. people, organizations, places and events) are extracted and represented by means of concept maps. Finally, these results: a) provide support to government bodies "end-users"; b) can be included in a GIS; c) can constitute the basis for the application of Artificial Adaptive Systems used in the domain of semantic analysis of
Uploads
Papers by Francesco Gargiulo