Speaker "Khosrow Hassibi" Details Back



Prediction of tonnage for WiFi Access Points for better placement and optimization


For Telcos, optimal WiFi access points’ (AP) placement is of great importance in terms of coverage to ensure good customer experience. A lot of planning is done to select the new AP locations and to optimize the existing APs’ usage. Big data infrastructures are already in place to collect and store session data for each AP, across all user devices. This big data has many different use cases from a data science perspective. There are different predictive metrics that can assist in planning and deployment of new APs and tuning of the existing ones. In this presentation, I explain what modeling approaches were taken into consideration using forecasting, classification, and estimation techniques and how they could be used to address the business operational problems of interest.


Dr. Hassibi is an expert, a pioneer, practitioner, and thought leader in the areas of data science, machine learning, and AI. His expertise is based on twenty plus years of design, R&D, consulting/sales, and management in applying these technologies to hard real-world business problems such as real-time fraud detection, hand-print OCR, robotics, marketing, risk, preventive maintenance, and transactional customer behavior analysis. Dr. Hassibi is recognized for his contributions to real-time payment card fraud detection (FalconTM) and has been a part of four machine learning startups focused on new data products and analytics-based business solutions. Most recently and prior to joining R Systems as Chief Data Scientist, he has been with SAS Institute and Altice USA (formerly Cablevision) with his main interest focused on new big data analytics and machine learning applications in financial and telecommunication industries. His has a recent book on machine learning and big data titled "High performance Data Mining and Big Data Analytics." (