Clustering Interest Graphs for Customer Segmentation Problems

Abstract

We propose the clustering-based approach for customer segmentation based on their digital traces. The weighted graph of co-occurrences of interests is created (with and without an account of socio-demographical stratum), and the graph is further clustered with Markov-based or modularity-based approaches. The method's work is illustrated using real data about customer activities on the Internet provided by our partner. We study and discuss segmentation results for different approaches and further applications for the development of marketing products for segments identified. Both clustering methods allow for identifying consistent interest segments which may be used for planning cross-sell campaigns as well as for personalized product suggestions.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 6 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.