A5026005294- Data Mining Algorithms and Applications
- Data Stream Mining Techniques
- Network Security and Intrusion Detection
- Machine Learning and Data Classification
- Context-Aware Activity Recognition Systems
- Anomaly Detection Techniques and Applications
- Manufacturing Process and Optimization
- Internet Traffic Analysis and Secure E-voting
- Imbalanced Data Classification Techniques
- Healthcare Technology and Patient Monitoring
- Explainable Artificial Intelligence (XAI)
- Topic Modeling
- Industrial Vision Systems and Defect Detection
- Energy Efficient Wireless Sensor Networks
- Embedded Systems Design Techniques
- Fire Detection and Safety Systems
- Data Management and Algorithms
- Energy Load and Power Forecasting
- Fire effects on ecosystems
- Smart Grid Energy Management
- Time Series Analysis and Forecasting
- Genetic and Kidney Cyst Diseases
- Forecasting Techniques and Applications
- Building Energy and Comfort Optimization
- Face and Expression Recognition
Polytechnic University of Turin
2015-2024
Weatherford College
2021
Flint Institute Of Arts
2021
Turin Polytechnic University
2021
Forecasting has always been at the forefront of decision making and planning. The uncertainty that surrounds future is both exciting challenging, with individuals organisations seeking to minimise risks maximise utilities. large number forecasting applications calls for a diverse set methods tackle real-life challenges. This article provides non-systematic review theory practice forecasting. We provide an overview wide range theoretical, state-of-the-art models, methods, principles,...
This paper presents a flexible framework that performs real-time analysis of physiological data to monitor people's health conditions in any context (e.g., during daily activities, hospital environments). Given historical data, different behavioral models tailored specific particular disease, patient) are automatically learnt. A suitable model for the currently monitored patient is exploited stream classification phase. The has been designed perform both instantaneous evaluation and over...
The recent expansion of IoT-enabled (Internet Things) devices in manufacturing contexts and their subsequent data-driven exploitation paved the way to advent Industry 4.0, promoting a full integration IT services, smart devices, control systems with physical objects, electronics sensors. real-time transmission analysis collected data from factories has potential create intelligence, which predictive maintenance is an expression. Hence need design new approaches able manage not only volume,...
The overwhelming increase of parcel transports has prompted the need for effective and scalable intelligent logistics systems. In parallel, with advent Industry 4.0, a tight integration Internet Things technologies Big Data analytics solution become necessary to effectively manage industrial processes early predict product faults or service disruptions. context good transports, development smart monitoring tools is particularly useful couriers ensure efficient deliveries. However, existing...
Understanding the behavior of a network from large scale traffic dataset is challenging problem. Big data frameworks offer scalable algorithms to extract information raw data, but often require sophisticated fine-tuning and detailed knowledge machine learning algorithms. To streamline this process, we propose self-learning insightful analyzer (SeLINA), generic, self-tuning, simple tool measurements. SeLINA includes different analytics techniques providing capabilities state-of-the-art...
The pervasive and increasing deployment of smart meters allows collecting a huge amount fine-grained energy data in different urban scenarios. analysis such is challenging opening up variety interesting new research issues across computer science areas. key role scientists providing researchers practitioners with cutting-edge scalable analytics engines to effectively support their daily activities, hence fostering leveraging data-driven approaches. This paper presents SPEC, distributed...
Wildfires are one of the natural hazards that European Union is actively monitoring through Copernicus EMS Earth observation program which continuously releases public information related to such catastrophic events. Such occurrences cause both short- and long-term damages. Thus, limit their impact plan restoration process, a rapid intervention by authorities needed, can be enhanced use satellite imagery automatic burned area delineation methodologies, accelerating response decision-making...
Energy efficiency and energy consumption awareness are a growing priority for many countries. Among the large variety of methods proposed by scientists professionals to evaluate building consumption, widely adopted approach is signature. Since data easily scale towards very datasets, problem characterizing through signature from these huge collections becomes challenging. This paper presents distributed system, named ESA, collection, storage, analysis amount energy-related keep continuously...
Large volumes of data are being produced by various modern applications at an ever increasing rate. These range from wireless sensors networks to social networks. The automatic analysis such huge volume is a challenging task since large amount interesting knowledge can be extracted. Association rule mining exploratory method able discover and hidden correlations among data. Since this process characterized computationally intensive tasks, efficient distributed approaches needed increase its...
Summary Public genomic and proteomic databases can be affected by a variety of errors. These errors may involve either the description or meaning data (namely, syntactic semantic errors). We focus our analysis on detection errors, in order to verify accuracy stored information. In particular, we address issue constraints functional dependencies among attributes given relational database. Constraints show semantics database schema their knowledge exploited improve quality integration design,...
Selecting a small number of discriminative genes from thousands is fundamental task in microarray data analysis. An effective feature selection allows biologists to investigate only subset instead the entire set, thus avoiding insigni
In recent years, the number of industry-4.0-enabled manufacturing sites has been continuously growing, and both quantity variety signals data collected in plants are increasing at an unprecedented rate. At same time, demand Big Data processing platforms analytical tools tailored to environments become more prominent. Manufacturing companies collecting huge amounts information during production process through a plethora sensors networks. To extract value actionable knowledge from such...
In the last few years, a large number of smart meters have been deployed in buildings to continuously monitor fine-grained energy consumption. Meteorological data deeply impact consumption, and an in-depth analysis collected correlated can uncover interesting actionable insights improve overall balance our communities enhance people’s awareness wasting. To effectively extract meaningful interpretable from collections measurements multi-dimensional meteorological data, innovative science...
Abstract Despite the high accuracy offered by state-of-the-art deep natural-language models (e.g., LSTM, BERT), their application in real-life settings is still widely limited, as they behave like a black-box to end-user. Hence, explainability rapidly becoming fundamental requirement of future-generation data-driven systems based on deep-learning approaches. Several attempts fulfill existing gap between and interpretability have been made. However, robust specialized eXplainable Artificial...
The ability to correctly identify areas damaged by forest wildfires is essential plan and monitor the restoration process estimate environmental damages after such catastrophic events. wide availability of satellite data, combined with recent development machine learning deep methodologies applied computer vision field, makes it extremely interesting apply aforementioned techniques field automatic burned area detection. One main issues in a context limited amount labeled especially semantic...
Frequent closed itemset mining is among the most complex exploratory techniques in data mining, and provides ability to discover hidden correlations transactional datasets. The explosion of Big Data leading new parallel distributed approaches. Unfortunately, them are designed cope with low-dimensional datasets, whereas no highdimensional frequent algorithms exists. This work introduces PaMPa-HD, a MapReduce-based algorithm for high-dimensional based on Carpenter. experimental results,...
Automating predictive machine learning entails the capability of properly triggering update trained models. To this aim, degradation models has to be continuously evaluated over time detect data distribution drifts between original training set and new data. Traditionally, prediction performance is used as a metric. However, quality indices require ground-truth class labels known for newly classified data, making them unsuitable real-time applications, might totally absent or available only later.