Luca Venturini
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Luca Venturini has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
Elena Baralis
Department of Control and Computer Engineering Politecnico di Torino Corso Duca degli Abruzzi 24 Torino Italy
5 shared publicationsPaolo Garza
Department of Control and Computer Engineering Politecnico di Torino Corso Duca degli Abruzzi 24 Torino Italy
4 shared publicationsTania Cerquitelli
4 shared publicationsDaniele Apiletti
Politecnico di Torino
2 shared publications- Daniele Apiletti
Department of Control and Computer Engineering, Politecnico di Torino, 10129 Turin, Italy
2 shared publications - Marco Mellia
Politecnico di Torino, Torino, Italy
2 shared publications
Latest Publications
- Open Access
Scaling associative classification for very large datasets
Supervised learning algorithms are nowadays successfully scaling up to datasets that are very large in volume, leveraging the potential of in-memory cluster-computing Big Data frameworks. Still, massive datasets with a number of large-domain categorical features are a difficult challenge for any classifier. Most off-the-shelf solutions cannot cope with this problem. In this work we introduce DAC, a Distributed Associative Classifier. DAC exploits ensemble learning to distribute the training of an associative classifier among parallel workers and improve the final quality of the model. Furthermore, it adopts several novel techniques to reach high scalability without sacrificing quality, among which a preventive pruning of classification rules in the extraction phase based on Gini impurity. We ran experiments on Apache Spark, on a real large-scale dataset with more than 4 billion records and 800 million distinct categories. The results showed that DAC improves on a state-of-the-art solution in both prediction quality and execution time. Since the generated model is human-readable, it can not only classify new records, but also allow understanding both the logic behind the prediction and the properties of the model, becoming a useful aid for decision makers.
Open AccessScaling associative classification for very large datasets
Supervised learning algorithms are nowadays successfully scaling up to datasets that are very large in volume, leveraging the potential of in-memory cluster-computing Big Data frameworks. Still, massive datasets with a number of large-domain categorical features are a difficult challenge for any classifier. Most off-the-shelf solutions cannot cope with this problem. In this work we introduce DAC, a Distributed Associative Classifier. DAC exploits ensemble learning to distribute the training of an associative classifier among parallel workers and improve the final quality of the model. Furthermore, it adopts several novel techniques to reach high scalability without sacrificing quality, among which a preventive pruning of classification rules in the extraction phase based on Gini impurity. We ran experiments on Apache Spark, on a real large-scale dataset with more than 4 billion records and 800 million distinct categories. The results showed that DAC improves on a state-of-the-art solution in both prediction quality and execution time. Since the generated model is human-readable, it can not only classify new records, but also allow understanding both the logic behind the prediction and the properties of the model, becoming a useful aid for decision makers.
- Open Access
- 8 December 2017
- Published by Springer Nature in Journal of Big Data
- Vol. 4 (1) , 44
Scaling associative classification for very large datasets
- 8 December 2017
- Published by Springer Nature in Journal of Big Data
- Vol. 4 (1) , 44
Open AccessScaling associative classification for very large datasets
- Open Access
- 1 December 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3779-3783
Analyzing spatial data from twitter during a disaster
Social media can be an invaluable help in a mass emergency, but the information handling can be challenging. One major concern is identifying posts related to the area, or pinning them on a map. This exploratory study analyzes the spatial data coming with tweets during two natural disasters, an earthquake and a hurricane. Geo-tagged tweets confirm to be a small fraction of all tweets and disasters within a limited region appear to be a niche topic in the whole stream. The results can help researchers and practitioners in the design of tools to identify these messages.
- 1 December 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3779-3783
Open AccessAnalyzing spatial data from twitter during a disaster
Social media can be an invaluable help in a mass emergency, but the information handling can be challenging. One major concern is identifying posts related to the area, or pinning them on a map. This exploratory study analyzes the spatial data coming with tweets during two natural disasters, an earthquake and a hurricane. Geo-tagged tweets confirm to be a small fraction of all tweets and disasters within a limited region appear to be a niche topic in the whole stream. The results can help researchers and practitioners in the design of tools to identify these messages.
- Open Access
- 1 September 2017
- Published by Elsevier in Big Data Research
- Vol. 9, 67-83
Frequent Itemsets Mining for Big Data: A Comparative Analysis
- 1 September 2017
- Published by Elsevier in Big Data Research
- Vol. 9, 67-83
Open AccessFrequent Itemsets Mining for Big Data: A Comparative Analysis
- 1 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Performance Comparison and Optimization of ICN Prototypes
Information Centric Networking paradigm is a solid proposal to update the original Internet design. While the research community is still working on technical improvement, the technology has reached a stage that allows the design, testing and deployment of ICN specific applications. To this goal, the development of testbeds allowing researchers to evaluate their applications in a real setup is of key importance. There are few existing ICN prototypes that would allow experimenters to work in this direction. Their performance is fundamental, so that they do not create artificial bottlenecks affecting test reliability. Unfortunately, existing solutions have been designed to demonstrate the feasibility of the ICN solution, thus lacking of optimization and scalability required to run reliable scalability tests. In this paper we address this problem by first, offering a performance comparison between available ICN prototypes and, second by proposing a revisited system design, CCN-par, to increase scalability, and remove bottlenecks that hamper current software-based testbeds.
- 1 December 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Performance Comparison and Optimization of ICN Prototypes
Information Centric Networking paradigm is a solid proposal to update the original Internet design. While the research community is still working on technical improvement, the technology has reached a stage that allows the design, testing and deployment of ICN specific applications. To this goal, the development of testbeds allowing researchers to evaluate their applications in a real setup is of key importance. There are few existing ICN prototypes that would allow experimenters to work in this direction. Their performance is fundamental, so that they do not create artificial bottlenecks affecting test reliability. Unfortunately, existing solutions have been designed to demonstrate the feasibility of the ICN solution, thus lacking of optimization and scalability required to run reliable scalability tests. In this paper we address this problem by first, offering a performance comparison between available ICN prototypes and, second by proposing a revisited system design, CCN-par, to increase scalability, and remove bottlenecks that hamper current software-based testbeds.
- 14 August 2016
- Published by Springer Nature
- p. 137-146
BAC: A Bagged Associative Classifier for Big Data Frameworks
- 14 August 2016
- Published by Springer Nature
- p. 137-146
BAC: A Bagged Associative Classifier for Big Data Frameworks
- 2 August 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 13 (3) , 696-710
SeLINA: A Self-Learning Insightful Network Analyzer
Understanding the behavior of a network from a large scale traffic dataset is a challenging problem. Big data frameworks offer scalable algorithms to extract information from raw data, but often require a sophisticated fine-tuning and a detailed knowledge of machine learning algorithms. To streamline this process, we propose self-learning insightful network analyzer (SeLINA), a generic, self-tuning, simple tool to extract knowledge from network traffic measurements. SeLINA includes different data analytics techniques providing self-learning capabilities to state-of-the-art scalable approaches, jointly with parameter auto-selection to off-load the network expert from parameter tuning. We combine both unsupervised and supervised approaches to mine data with a scalable approach. SeLINA embeds mechanisms to check if the new data fits the model, to detect possible changes in the traffic, and to, possibly automatically, trigger model rebuilding. The result is a system that offers human-readable models of the data with minimal user intervention, supporting domain experts in extracting actionable knowledge and highlighting possibly meaningful interpretations. SeLINA's current implementation runs on Apache Spark. We tested it on large collections of real-world passive network measurements from a nationwide ISP, investigating YouTube, and P2P traffic. The experimental results confirmed the ability of SeLINA to provide insights and detect changes in the data that suggest further analyses.
- 2 August 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 13 (3) , 696-710
SeLINA: A Self-Learning Insightful Network Analyzer
Understanding the behavior of a network from a large scale traffic dataset is a challenging problem. Big data frameworks offer scalable algorithms to extract information from raw data, but often require a sophisticated fine-tuning and a detailed knowledge of machine learning algorithms. To streamline this process, we propose self-learning insightful network analyzer (SeLINA), a generic, self-tuning, simple tool to extract knowledge from network traffic measurements. SeLINA includes different data analytics techniques providing self-learning capabilities to state-of-the-art scalable approaches, jointly with parameter auto-selection to off-load the network expert from parameter tuning. We combine both unsupervised and supervised approaches to mine data with a scalable approach. SeLINA embeds mechanisms to check if the new data fits the model, to detect possible changes in the traffic, and to, possibly automatically, trigger model rebuilding. The result is a system that offers human-readable models of the data with minimal user intervention, supporting domain experts in extracting actionable knowledge and highlighting possibly meaningful interpretations. SeLINA's current implementation runs on Apache Spark. We tested it on large collections of real-world passive network measurements from a nationwide ISP, investigating YouTube, and P2P traffic. The experimental results confirmed the ability of SeLINA to provide insights and detect changes in the data that suggest further analyses.
- 1 April 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 812-816
SaFe-NeC: A scalable and flexible system for network data characterization
Nowadays, large volumes of data and measurements are being continuously generated by computer and telecommunication networks, but such volumes make it difficult to extract meaningful knowledge from them. This paper presents SaFe-NeC, an innovative methodology for analyzing network traffic by exploiting data mining techniques, i.e. clustering and classification algorithms, focusing on self-learning capabilities of state-of-the-art scalable approaches. Self-learning algorithms, coupled with self-assessment indicators and domain-driven semantics enriching data mining results, are able to build a model of the data with minimal user intervention and highlight possibly meaningful interpretations to domain experts. Furthermore, a self-evolving model evaluation phase is included to continuously track the quality degradation of the model itself, whose rebuilding is triggered as soon as quality indicators fall below a threshold of tolerance. The proposed methodology can exploit the computational advantages of distributed computing frameworks, as the current implementation runs on Apache Spark. Preliminary experimental results on a real traffic dataset show the full potential of the proposed methodology to characterize network traffic data.
- 1 April 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 812-816
SaFe-NeC: A scalable and flexible system for network data characterization
Nowadays, large volumes of data and measurements are being continuously generated by computer and telecommunication networks, but such volumes make it difficult to extract meaningful knowledge from them. This paper presents SaFe-NeC, an innovative methodology for analyzing network traffic by exploiting data mining techniques, i.e. clustering and classification algorithms, focusing on self-learning capabilities of state-of-the-art scalable approaches. Self-learning algorithms, coupled with self-assessment indicators and domain-driven semantics enriching data mining results, are able to build a model of the data with minimal user intervention and highlight possibly meaningful interpretations to domain experts. Furthermore, a self-evolving model evaluation phase is included to continuously track the quality degradation of the model itself, whose rebuilding is triggered as soon as quality indicators fall below a threshold of tolerance. The proposed methodology can exploit the computational advantages of distributed computing frameworks, as the current implementation runs on Apache Spark. Preliminary experimental results on a real traffic dataset show the full potential of the proposed methodology to characterize network traffic data.
- 1 April 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 112-116
Monitoring the citizens' perception on urban security in Smart City environments
Sensing the perception of citizens on urban security is a key point in Smart City management. To address non-emergency issues municipalities commonly acquire citizens' reports and then analyze them offline to perform targeted actions. However, since non-emergency data potentially scale towards Big Data there is a need for open standards and technologies to enable data mining and Business Intelligence analyses. The paper presents an integrated data mining and Business Intelligence architecture, relying on open technologies, for the analysis of non-emergency open data acquired in a Smart City context. Non-emergency data are first enriched with additional information related to the context of the warning reports and then analyzed offline to generate (i) informative dashboards based on a selection of Key Performance Indicators (KPIs), and (iii) association rules representing implications between warning categories and contextual information (e.g., city areas, districts, time slots). KPIs and rules are exploited to selectively notify to municipality actors (assessors, area operators) potentially critical situations, according to their role and authority. The experiments demonstrate the effectiveness of the proposed approach in a real Smart City context.
- 1 April 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 112-116
Monitoring the citizens' perception on urban security in Smart City environments
Sensing the perception of citizens on urban security is a key point in Smart City management. To address non-emergency issues municipalities commonly acquire citizens' reports and then analyze them offline to perform targeted actions. However, since non-emergency data potentially scale towards Big Data there is a need for open standards and technologies to enable data mining and Business Intelligence analyses. The paper presents an integrated data mining and Business Intelligence architecture, relying on open technologies, for the analysis of non-emergency open data acquired in a Smart City context. Non-emergency data are first enriched with additional information related to the context of the warning reports and then analyzed offline to generate (i) informative dashboards based on a selection of Key Performance Indicators (KPIs), and (iii) association rules representing implications between warning categories and contextual information (e.g., city areas, districts, time slots). KPIs and rules are exploited to selectively notify to municipality actors (assessors, area operators) potentially critical situations, according to their role and authority. The experiments demonstrate the effectiveness of the proposed approach in a real Smart City context.