Papers by Fabio Schreiber
Big Data and Cognitive Computing
One of the travelers’ main challenges is that they have to spend a great effort to find and choos... more One of the travelers’ main challenges is that they have to spend a great effort to find and choose the most desired travel offer(s) among a vast list of non-categorized and non-personalized items. Recommendation systems provide an effective way to solve the problem of information overload. In this work, we design and implement “The Hybrid Offer Ranker” (THOR), a hybrid, personalized recommender system for the transportation domain. THOR assigns every traveler a unique contextual preference model built using solely their personal data, which makes the model sensitive to the user’s choices. This model is used to rank travel offers presented to each user according to their personal preferences. We reduce the recommendation problem to one of binary classification that predicts the probability with which the traveler will buy each available travel offer. Travel offers are ranked according to the computed probabilities, hence to the user’s personal preference model. Moreover, to tackle th...
Abstract. The formidable amount of heterogeneous information, accessed by the networked enterpris... more Abstract. The formidable amount of heterogeneous information, accessed by the networked enterprise through all the available channels, makes it difficult for users to find the right information at the right time and at the right level of detail. Contextual meta-data about the system and the users can be used to reduce this plethora of information, providing high-quality, focussed knowledge to users and applications at all decision-making points. We propose context-aware system design methodologies and techniques exemplified within the wine production scenario, where several classes of users access the networked-enterprise data sources, the sensors used for monitoring the productive cycle, and external sources of different nature.
Pushing context-awareness down to the core: more
Information Processing & Management, 1984
Big Data and Cognitive Computing, 2022
One of travelers’ main challenges is that they have to spend a great effort to find and choose th... more One of travelers’ main challenges is that they have to spend a great effort to find and choose the most desired travel offer(s) among a vast list of non-categorized and non-personalized items.
Recommendation systems provide an effective way to solve the problem of information overload. In this work, we design and implement “The Hybrid Offer Ranker” (THOR), a hybrid, personalized recommender system for the transportation domain. THOR assigns every traveler a unique contextual preference model built using solely their personal data, which makes the model sensitive to the user’s choices. This model is used to rank travel offers presented to each user according to their personal preferences. We reduce the recommendation problem to one of binary classification that predicts the probability with which the traveler will buy each available travel offer. Travel offers are ranked according to the computed probabilities, hence to the user’s personal preference model.
Moreover, to tackle the cold start problem for new users, we apply clustering algorithms to identify groups of travelers with similar profiles, and build a preference model for each group. To test the system’s performance, we generate a dataset according to some carefully designed rules.
The results of the experiments show that the THOR tool is capable of learning the contextual preferences of each traveler and ranks offers starting from those that have the higher probability of being selected
Edge-of-Things in Personalized Healthcare Support Systems, 2022
The emergence of Context-aware systems in the domains of autonomic, monitoring, and safety-critic... more The emergence of Context-aware systems in the domains of autonomic, monitoring, and safety-critical applications asks for the definition of methods to formally assess their correctness and dependability properties. Many of these properties are common to Automatic Control systems, a field that developed well established analysis and design techniques to formalize and investigate them. In this paper, we use Boolean Control Networks, to discuss some properties of a feedback Context-aware system in a case study based on a healthcare management example.
The formidable amount of heterogeneous information, accessed by the networked enterprise through ... more The formidable amount of heterogeneous information, accessed by the networked enterprise through all the available channels, makes it difficult for users to find the right information at the right time and at the right level of detail. Contextual meta-data about the system and the users can be used to reduce this plethora of information, providing high-quality, focussed knowledge to users and applications at all decision-making points. We propose context-aware system design methodologies and techniques exemplified within the wine production scenario, where several classes of users access the networked-enterprise data sources, the sensors used for monitoring the productive cycle, and external sources of different nature
Journal on Computing and Cultural Heritage, 2017
Information and Communication Technologies have radically changed the modern Cultural Heritage sc... more Information and Communication Technologies have radically changed the modern Cultural Heritage scenery: Simple traditional Information Systems supporting the management of cultural artifacts have left the place to complex systems that expose rich information extracted from heterogeneous data sources—like Sensor Networks, Social Networks, Digital Libraries, Multimedia Collections, Web Data Service, and so on—by means of sophisticated applications that enhance the users’ experience. In this article, we describe SCRABS, a Smart Context-awaRe Browsing assistant for cultural EnvironmentS . SCRABS has been developed during the Cultural Heritage Information Systems national project and promoted by DATABENC, the Cultural Heritage Technological District of the Campania Region, in Italy. SCRABS has been designed on top of a Big Data technological stack as the result of a multidisciplinary project carried out by a heterogeneous team of computer scientists, archeologists, architects, and expert...
international journal of next-generation computing, Dec 12, 2019
Electric Vehicle Sharing Services for Smarter Cities, 2017
The creation of intelligent pervasive spaces is one of the most interesting opportunities offered... more The creation of intelligent pervasive spaces is one of the most interesting opportunities offered by pervasive systems: social and physical ambients can be created with the aid of ICT technologies, providing enhanced capabilities for humans to interact with the surrounding environment. In general, these features are useful for providing security services, energy management, water and pollution control or to create assisted-living ambients for impaired or elderly people, but constitute also proactive and intelligent supports to novel applications in traffic management.
Page 1. A Model for the Human Computer Interface Evaluation in Safety Critical Computer Applicati... more Page 1. A Model for the Human Computer Interface Evaluation in Safety Critical Computer Applications Fabio A. Schreiber Dipartimento di Elettronica e Informazione Politecnico di Milano Proceedings of the IEEE International ...
Euromicro Newsletter, 1975
ABSTRACT In this paper an interface device for computer communication is described, based on the ... more ABSTRACT In this paper an interface device for computer communication is described, based on the use of a programmable microprocessor. The aim of this Micro Front End Processor is to free the main computer from all the communication procedures management, as if some intelligence were added to the transmission line. An outline of the model, the realization of the interlocutor (the entity carrying on the colloquy) is based on, is given, the heart of the device being a software implemented Variable Structure sequential Machine. Hardwired and microprogrammed versus microcomputerized realization tradeoff is examined with respect to cost, performance, and flexibility, and the choosing criteria among the possible architectures and technological features of microcomputers are reviewed. Finally, a block diagram and a functional description of a device, based on Intel 8080 microprocessor, implementing the colloquy for the UNIVAC 1108 under EXEC 8 operating system are presented as well as an estimate of the performance of such a device.
NATO ASI Series, 1994
I firmly trust in the need for formal methods in specification and in verification/validation of ... more I firmly trust in the need for formal methods in specification and in verification/validation of real-time systems to get 100 % confidence in their performance and dependability, especially when safety is a main concern. However, I still see a very long way to go before they are widely accepted and used by the practitioners community. There are several reasons for that: 1. cultural — our believes notwithstanding, most of real-time programmers are still assembler programmers the same way as most of commercial application programmers are Cobol programmers, and scientific application programmers are Fortran programmers (if you don’t believe it, just scrap the surface and see!), so they are not ready to accept and understand the often awful mathematical notation and machinery required for formal techniques; 2. practical — at the state-of-art, only small and well understood pieces of the whole system can be treated formally, so let us concentrate on the most critical parts and leave the rest to the traditional methods. Moreover, formal techniques are applied to source code; they should be applied also to prove the correctness of the compiler and of other tools (e.g. mathematical libraries), not only to the real-time application software; 3. structural — starting from low level program verification, formal methods have been pushed up to the specification level. However, we cannot leave statistical dependability assessment techniques totally off the game since formal specification again needs validation against the designer original thought; therefore there will always be a passage point from informal to formal and the risk of making formally undetectable errors is there![1]
Decision Support Systems, 1989
The problem of document management in an Office Information System in view of supporting the deci... more The problem of document management in an Office Information System in view of supporting the decision process by a set of users with different knowledge, capabilities and requirements is considered. User profiles are used to improve the retrieval properties and a control system is outlined to automatically update the utility information used by the searching strategies, user requirements are modeled in Dynamic User Profiles in order to maintain accurate knowledge of the most promising approach to document search, of the interest fields, and of the favourite output format. An example of DUP use is provided adopting a flexible query specification to take advantage of database selection, retrieval via key'word, or full text scanning in a multi-strategy query processing approach. Actual queries are redefined by extracting the aspects suitable for each strategy.
A survey of context models and a preliminary methodology for context driven data view definition
The RECKON project focuses on interconnection technologies and context-aware data-analytics techn... more The RECKON project focuses on interconnection technologies and context-aware data-analytics techniques to improve safety in workplaces, with the ultimate objective of identifying and preventing dangerous situations before accidents occur. In RECKON, prevention is interpreted through the latest monitoring, diagnostics and prognostics techniques from a safety perspective, allowing to detect and use, even in real time, a large amount of data about the entire operational context. Using sensor networks, we are able to collect information that is used in two ways: (i) when a potentially dangerous situation is detected, the system raises an alarm to prevent an accident, and (ii) whenever an accident or a near-miss (i.e., a potential accident that was narrowly averted) occurs, the related useful information is stored in a case report automatically generated and later used to update the accident-prevention politics. This work briefly describes the operational framework of RECKON, along with ...
Sistemi Evoluti per Basi di Dati, 2004
Uploads
Papers by Fabio Schreiber
Recommendation systems provide an effective way to solve the problem of information overload. In this work, we design and implement “The Hybrid Offer Ranker” (THOR), a hybrid, personalized recommender system for the transportation domain. THOR assigns every traveler a unique contextual preference model built using solely their personal data, which makes the model sensitive to the user’s choices. This model is used to rank travel offers presented to each user according to their personal preferences. We reduce the recommendation problem to one of binary classification that predicts the probability with which the traveler will buy each available travel offer. Travel offers are ranked according to the computed probabilities, hence to the user’s personal preference model.
Moreover, to tackle the cold start problem for new users, we apply clustering algorithms to identify groups of travelers with similar profiles, and build a preference model for each group. To test the system’s performance, we generate a dataset according to some carefully designed rules.
The results of the experiments show that the THOR tool is capable of learning the contextual preferences of each traveler and ranks offers starting from those that have the higher probability of being selected
Recommendation systems provide an effective way to solve the problem of information overload. In this work, we design and implement “The Hybrid Offer Ranker” (THOR), a hybrid, personalized recommender system for the transportation domain. THOR assigns every traveler a unique contextual preference model built using solely their personal data, which makes the model sensitive to the user’s choices. This model is used to rank travel offers presented to each user according to their personal preferences. We reduce the recommendation problem to one of binary classification that predicts the probability with which the traveler will buy each available travel offer. Travel offers are ranked according to the computed probabilities, hence to the user’s personal preference model.
Moreover, to tackle the cold start problem for new users, we apply clustering algorithms to identify groups of travelers with similar profiles, and build a preference model for each group. To test the system’s performance, we generate a dataset according to some carefully designed rules.
The results of the experiments show that the THOR tool is capable of learning the contextual preferences of each traveler and ranks offers starting from those that have the higher probability of being selected