Phone support
+46 70 733 36 34
Email us
info@redfield.ai
Address
Flemminggatan 15A 112 26 Stockholm Sweden
Case Study Recommendation Engines
Client
Allente
Industry
Broadcast media
About the Company
Allente is a Nordic TV distribution company offering high-quality satellite, streaming services, IPTV solutions, and fiber broadband to more than one million customers in Sweden, Norway, Denmark, and Finland.
Opportunity
The client was looking for a platform that would recommend to the customers movies, series, or shows based on their view history and personal preferences. The solution was implemented on Knime Server.
The requirements for the platform were the following:
With the implementation of a data science & machine learning platform, the client aimed to capitalize on the goals set out:
The data science & machine learning platform will be a central tool for analysis, especially in the Customer Service department. Other departments will also be able to use the tool – depending on what insights the team aims to get.
How Did Our Solution Work
Step 1
On-boarding and installation
The first objective was to get our client’s team familiar with Knime through training and exercises.
The training included:
Step 2
Data Exploration
The objective of this phase was to explore the data and report to Allente any data gaps or any issues. Redfield has been analyzing three data sources provided by the client: event data, a video-on-demand catalog, and sample tables from its legacy Oracle data. This step was an initial milestone to ensure access to the data sources, the setup of the initial Knime workflows, and ensure smooth delivery of the subsequent phases.
Step 3
Content Recommendation
Based on the data explored, the data content recommendation model was implemented in the following steps:
Why Choose Us
Result
We collected and analyzed enough data to build a recommendation model with partial coverage. The model would be able to better understand the content that users are engaged in
(vs content they only partially watched) if we can use time spent watching an asset to distinguish between the use cases.
We will also be able to deliver recommendations to
more users using such metrics as page views to a particular piece of content. This gave us the basis to recommend content to people for whom we don’t have any viewing
history.
The goal was to create a reference implementation so that Allente data
science team could enhance as new consumption data surfaces. We use the optimal set of open-source tools to deliver the best results to your project. Our solution is no-code, easy to implement and adapt. At the same time, it is fast and flexible to implement.
Tools
Technologies Used
Knime Analytics Platform
Knime Server
Redfield Managed Service
TESTIMONIALS
Client's Experience
For us it has been key to have one tool that could do a little bit of everything. Easy to use – yet advanced and powerful. Having one hub where we can do both data-crunching, machine learning models and automated pipelines in a simple manner has transformed our working life.
