A5075371008- Big Data and Business Intelligence
- Scientific Computing and Data Management
- Metaheuristic Optimization Algorithms Research
- Vehicle Routing Optimization Methods
- Simulation Techniques and Applications
- Manufacturing Process and Optimization
- Face and Expression Recognition
- Neural Networks and Applications
- Service-Oriented Architecture and Web Services
- Image Retrieval and Classification Techniques
- Research Data Management Practices
- Fire effects on ecosystems
- Customer churn and segmentation
- Advanced Manufacturing and Logistics Optimization
- Fire Detection and Safety Systems
- Landslides and related hazards
- Auction Theory and Applications
- Optimization and Search Problems
- Big Data Technologies and Applications
- Imbalanced Data Classification Techniques
- Semantic Web and Ontologies
- Data Mining Algorithms and Applications
- Spreadsheets and End-User Computing
- Context-Aware Activity Recognition Systems
- Computational Physics and Python Applications
University of Trento
2018-2020
CMCC Foundation - Euro-Mediterranean Center on Climate Change
2013-2015
Central Maine Community College
2013
In feature selection, a measure that captures nonlinear relationships between features and class is the mutual information (MI), which based on how in reduces uncertainty output. this paper, we propose new related to MI, called neighborhood entropy, novel filter method its minimization greedy procedure. Our algorithm integrates sequential forward selection with approximated nearest-neighbors techniques locality-sensitive hashing. Experiments show classification accuracy usually higher than...
The explosion of the Internet has deeply affected labour market.Identifying most rewarded and demanded items in job offers is key for recruiters candidates.This work analyses 4, 000 from a Spanish IT recruitment portal.We conclude that (1) experience more than education, (2) we identify five profile clusters based on required skills (3) develop an accurate salary-range classifier by using tree-based ensembles.
The availability of systems able to process and analyse big amount data has boosted scientific advances in several fields. Workflows provide an effective tool define manage large sets processing tasks. In the analytics area, Ophidia project provides a cross-domain framework for analysis scientific, multi-dimensional datasets. exploits server-side, declarative, parallel approach mining. It also features complete workflow management system support execution complex analysis, schedule tasks...
Prevention is one of the most important stages in wildfire and other natural hazard management. Fire Danger Rating Systems (FDRSs) have been adopted by many countries to enhance prevention suppression planning. With aim provide real-time fire danger forecasts finer-scale behaviour analysis, an operational platform has developed within OFIDIA project (Operational FIre preventIon plAtform). The System consists (1) a data archive for managing weather forecasting wireless sensors data, (2)...
In simulation-based optimization, the optimal setting of input parameters objective function can be determined by heuristic optimization techniques. However, when simulators model stochasticity real-world problems, their output is a random variable and multiple evaluations are necessary to properly compare expected performance different parameter settings. This paper presents novel reactive sample size algorithm based on parametric tests indifference-zone selection, which used for improving...