Скачать или смотреть Data Science and Machine Learning Approaches for E Commerce - MeasureSummit 2023 (by Balazs Vajna)

This video was shown on MeasureSummit 2023 as an on-demand session, giving a quick summary of data science / machine learning / statistical analysis (/ "AI", as we call it these days) methods that can be used by e-commerce (or other retail) companies.

After a high overview of analytics solutions / project "ideas" across the customer lifecycle, the first half of the video shows different types of modelling use cases: propensity modelling (e.g. cross-sell, churn), Marketing Mix Modelling, price elasticity, demand forecasting, market basket analysis, customer segmentation (clustering), dimensionality reduction / recommendation engines.

The second half of the video shows the various steps (stages) of a data science project:
building the training data for predictive models (population and target definition, feature engineering
variable selection / reduction (also called "EDA" i.e. exploratory data analysis - though it's not that exploratory here anymore)
pre-modelling data preparation (partitioning, class imbalance correction, outlier treatment etc)
model building
evaluation of models
model deployment (briefly)

If you need any assistance, reach out to me on LinkedIn (  / balazsvajna   ). Also check out GA4BigQuery.com (https://www.ga4bigquery.com/) for tangible use cases. (Not data science, but very useful stuff, some articles specifically for e-commerce!)