Customer Profile Identification and Change Prediction – Algorithmic identification of the usage created profile of customers and prediction of their behavior. The software gets and processes the historical usage data of customers and utilise

  • Statistical analysis (time-series analysis, A/Β testing, data categorization, trust intervals)
  • AI and Machine Learning (Decision Trees, Support Vector Machines, Linear Regression)
  • Neuron networks (LSTM, Convolutional Networks)
  • Big data stream processing

Αutomated Proposal Suggestion - Using Neurocom TariffSuite product, and its enhancements with AI algorithms and Bayesian statistics, AutoSprice identifies the best suitable offer for a customer. TariffSuite has been adapted to the processing pipeline of time-series data, utilizing the Apache Flink big data stream processing architecture. Then, using exploration/exploitation machine learning algorithms, it also identifies the right price, that is acceptable in terms of a. customer willingness to pay and b. company profitability criteria.

Commission Calculation System - A system component calculates in real-time the commission for the partner that is assigned for a specific sales action. The commission calculation algorithms of CommissionSuite platform of Neurocom, have been adapted to the big data stream processing and analytics platform of Autosprice, in order to: - sense and process all events that are relevant to any customer and relevant new proposal - immediately recall and use the available historical data that are required for commission recalculation

Visualisation of Results – Use of open-source data visualisation technology that realise dashboard for the monitoring overview of system and performance indicators of the proposed offers in real time, as well as the analysis of customer profile and behavior history.

We design efficient solutions for critical telecom BSS processes

Get Information about the Neurocom team, the working methodology and philosophy, and our innovative suggestions.