Κατεύθυνση Ψηφιακά Μέσα Επικοινωνίας και Περιβάλλοντα Αλληλεπίδρασης
Library of the Faculties of Political Science and Public Administration, Communication and Mass Media Studies, Turkish and Modern Asian Studies, Sociology

2015

2015

Βράχα Σταματίνα

Κωνσταντίνος Μουρλάς, Αναπληρωτής καθηγητής Τμήμα ΕΜΜΕ, ΕΚΠΑ

Αξιοποίηση δεδομένων μεγάλου όγκου και Προβλεπτική Ανάλυση στην ασφάλιση φωτοβολταϊκών κινδύνων

Greek

The aim of this report is to attempt an approach in scientific fields such as open data, analytics, big data, data mining techniques, machine learning, statistics, business intelligence and web search engines, especially in the insurance area.
In addition to this theoretical research, there will be a trial design and implementation of a big data analytics product with a focus on the insurance sector, specifically concerning the risk management of photovoltaics, either before or after the danger has happened, which will easily integrate data mining tools and the rest of existing company's IT infrastructure. This product will support features such as data mining techniques and predictive elements, dedicated to the insurance risk management and marketing concerning photovoltaics. It will definitely include bussiness intelligence systems which will worκ as advisory when connected with data of the company Mentor S.A. - Surveyors, valuers, engineers, concerning past damages in photovoltaics. The users of this platform will eventually have the opportunity to assess a probable damage in such an insurance field.
Another goal of such a trial version design, is to simplify the procedures of big data analytics into business processes of insurance companies. The proposed platform will provide a complete set of collaborative predictive analytics solutions which will enable the existing company's IT personnel to build useful structures of predictive models and score those models with huge data sets.
The Predictive Analytics Platform for Web-based Services is to offer a complete package of tools, supporting a wide variety of predictive analytical processes, which work against mostly big data sources, consisting of both structured and unstructured data. The platform is based on intelligent algorithms which will provide feasible solutions around the complexities associated with the predictive analytical modeling. The use of the platform will eventually help companies gain competitive advantages by understanding better the profiles of their clients and by simplifying the insurance claim handling procedure, which will result in significant savings in costs and time spent.

Social, Political and Economic sciences

Open Data, Big Data, Business Intelligence, Data Mining, Predictive Analytics, Machine Learning, Risk Assessment, Photovoltaics, Ιnsurance Fraud, User-Based-Insurance

No

0

Yes

58

117

Τοπική Ψηφιακή Βιβλιοθήκη

File access is restricted.

https://pergamos.lib.uoa.gr/uoa/dl/object/1447699