An Actionable Knowledge Discovery System in Regular Sports Services

Abstract

This work presents an actionable knowledge discovery system for real user needs with three steps. In the first step, it extracts and transforms existing data in the databases of the ERP and CRM systems of the sports facilities and loads them into a Data Warehouse. In a second phase, predictive models are applied to identify profiles more susceptible to abandonment. Finally, in the third phase, based on the previous models, experimental planning is carried out, with test and control groups, in order to find concrete actions for customer retention.

Extracting Actionable Knowledge to Increase Business Utility in Sport Services

“…This work extends previous works where this pipeline was presented (Pinheiro and Cavique 2018;Pinheiro and Cavique 2019). These works fill a gap in the study of retention through the use of machine learning techniques in regular sports services.…”

“…The proposed framework presented in (Cavique 2006;Pinheiro and Cavique 2018;Pinheiro and Cavique 2019) works in a multiple-step pipeline, shown in Table 6, where the columns phase, outcome, evaluation and groups are presented. The column 'groups' is presented in order to clarify and differentiate groups that have similar names but different meanings.…”

“…However, the implementation in real scenarios is still in development. We reuse the data pipeline presented in (Cavique 2006;Pinheiro and Cavique 2018;Pinheiro and Cavique 2019) with three steps: data collection, predictive model and loyalty actions. This data pipeline corresponds to the domain-driven approach presented by Cao (Cao et al 2007;Cao 2010), where the predictive model delivers actionable knowledge and the loyalty actions return a business utility measure.…”

“…In order to develop the experiment and test the results, we used data from a Lisbon sports facility to which we applied the Extraction, Transformation and Loading (ETL) processes as explained in previous works (Pinheiro and Cavique 2018;Pinheiro and Cavique 2019). Since the performance of Machine Learning techniques is obtained through the manipulation of values (Gama et al 2017), in addition to the attributes directly mapped from the source database, some attributes have been transformed, discretized through numeric-symbolic conversions, or created new attributes that derive from transformations made on the original data.…”

A bi‐objective procedure to deliver actionable knowledge in sport services

“…The same authors (Pinheiro & Cavique, 2019a) added actionable knowledge discovery with experimental planning, using test and control groups, in order to find and measure the effectiveness of actions aimed at customer retention. Subsequently, they added the business utility measure of a sequence of actions to prevent the withdrawal of users of regular sports services (Pinheiro & Cavique, 2019b).…”