Facebook Pixel

Artificial intelligence (AI) development for mobile network usage predictions

ipvision A/S

Company description

ipvision A/S delivers communication solutions to business users. We have approximately 2.500 customers and annual revenue of approximately 100 million DKK. We deliver many different integrations and communication solutions but primarily IP telephony, internet access, and mobile communications. 

Project description

When ipvision A/S makes a bulk purchase of minutes, data and SMS from large telecom companies we try to predict usage of our subscribers before they make a call, send SMS or use data. The better we are at estimating usage on a given day the better it is cost wise. 

Our current method is not good enough and we can see that the optimal choice of price plans per subscriber would lower our cost every month. We set an ambitious research goal to develop an AI algorithm that allows us to reduce the cost of approximate 40.000 DKK every month. There is also a possibility to setup the Machine Learning Feedback loop to make training and accuracy improvements continuous.
For the AI algorithm training and development, we will give access to 5 years of daily usage historical data from 10.000+ subscribers. We will also give information about the end-users price plan. The nature of the data and access to the raw dataset requires students to sign an NDA and work in an environment where the raw data is hosted on ipvision A/S premises.

We expect to use the algorithm afterwards in our daily operations and we will provide full support to make you succeed in the development. 

Student description

We expect a student that has a strong profile and experience with machine learning, algorithms modeling. Also, we prefer to have experience with databases and open sources tools. Students are expected to work remotely or in our headquarters - this would be agreed in the first weeks of a project start. 

Artificial intelligence (AI) development for mobile network usage predictions | Match My Thesis
Sep 15, 2019


Artificial intelligence


Machine Learning


Mobile Network

Big Data