Description:
a successfully conducted marketing campaign for advertising various types of products involves providing relevant advertising to the user of the Internet resource on which he is located, which in turn is associated with an accurate determination of the client’s category and his preferences. A thematic model of user behavior profiles within an Internet resource is built based on transactional data of interaction with the site from cookies.
Context:
Web service users need to be segmented to personalize service offerings. This can be done based on cookies, since the history of browsing pages indicates the interests of the user.
Solution:
building a segmentation model using topic modeling based on cookies in the form of a text description of the pages visited.
Results:
A pilot project was carried out and integrated into the customer’s business process:
To build the model we used:
Simulation results:
Client: Marketing agency
Technology stack: TopicNet, BigARTM, Python, gensim.
a successfully conducted marketing campaign for advertising various types of products involves providing relevant advertising to the user of the Internet resource on which he is located, which in turn is associated with an accurate determination of the client’s category and his preferences. A thematic model of user behavior profiles within an Internet resource is built based on transactional data of interaction with the site from cookies.
Context:
Web service users need to be segmented to personalize service offerings. This can be done based on cookies, since the history of browsing pages indicates the interests of the user.
Solution:
building a segmentation model using topic modeling based on cookies in the form of a text description of the pages visited.
Results:
A pilot project was carried out and integrated into the customer’s business process:
- The topics of 80% of the pages visited by the user were identified;
- 70% are interpretable;
- Prediction of socio-demographic parameters of the audience;
- The interests of 90% of visitors were identified based on the topics of the pages visited;
- 80% of interests are interpretable;
- Increased advertising campaign conversion by 10% during A/B testing;
- Changes have been made to 15% of ready-made advertising campaigns.
To build the model we used:
- List of URL pages visited by the user;
- Description of URL pages;
- Accompanying information about the user;
- The results of advertising shown to the user.
Simulation results:
- Behavior model of Internet resource users;
- Model for predicting the likelihood of response to an advertising campaign;
- Interpretation of the customer base.
Client: Marketing agency
Technology stack: TopicNet, BigARTM, Python, gensim.
