Topology-Based Clusterwise Regression for User Segmentation and Demand Forecasting

Rivera-Castro, Rodrigo; Pletnev, Aleksandr; Pilyugina, Polina; Diaz, Grecia; Nazarov, Ivan; Zhu, Wang; Burnaev, Evgeny

doi:10.1109/dsaa.2019.00048

Cited by 5 publications

(4 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, incorporating the 'voice of the customer' via a linguistic-based text-mining approach that combines both qualitative data from CX touchpoints and quantitative company data sources can help complement NPS insights (Zaki et al, 2016). Other promising algorithms include those designed to (a) analyze customers' qualitative NPS comments (Chant & Potter, 2019), (b) map comments to NPS classifications (Vanderheyden et al, 2019), and (c) cluster firms' user bases and plan customer demand at a granular level (Rivera-Castro et al, 2019). However, at present, evidence of enhanced productivity (for example) resulting from a link between NPS and Big Data analytics, remains scarce.…”

Section: Recommendationmentioning

confidence: 99%

The Net Promoter Score interrogated through a services marketing lens: Review and recommendations for service organizations

Lacohee,

Souchon,

Dickenson

et al. 2023

International Journal of Market Research

View full text Add to dashboard Cite

The Net Promoter Score (NPS) is used by many service organizations as a key customer loyalty metric and driver of growth. Yet, despite its widespread adoption and consequent benchmarking benefits, the methodological soundness and usefulness of NPS has been questioned. Compounding these problems, few guidelines on NPS enhancement strategies exist, since research on antecedents to NPS remains elusive. The combination of (a) questionable research underpinning the original proposal for NPS, and (b) an under-researched nomological net, creates an unfortunate paradox, whereby service organizations strive to raise a score that may or may not lead to organizational success, with little evidence-based knowledge on how to raise this score strategically. To address this, we utilize a guiding framework to interrogate NPS though a services marketing lens. We use this framework to underpin a critical review of NPS as it applies to service organizations, and through the review, identify and discuss ten NPS interrogations. We then propose recommendations that address each of these to enhance further knowledge development and improve ongoing practice of NPS by service organizations.

show abstract

Section: Recommendationmentioning

confidence: 99%

The Net Promoter Score interrogated through a services marketing lens: Review and recommendations for service organizations

Lacohee,

Souchon,

Dickenson

et al. 2023

International Journal of Market Research

View full text Add to dashboard Cite

show abstract

“…Similarly, [45] use a combination of time-series clustering and regression with an LSTM to forecast customer churn. Marketing tasks are a typical case for clustering time series in the industry, as [34] and [35] show.…”

Section: Clustering Of Time Seriesmentioning

confidence: 99%

COHORTNEY: Non-Parametric Clustering of Event Sequences

Zhuzhel,

Rivera-Castro,

Kaploukhaya

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

There is emerging attention towards working with event sequences. In particular, clustering of event sequences is widely applicable in domains such as healthcare, marketing, and finance. Use cases include analysis of visitors to websites, hospitals, or bank transactions. Unlike traditional time series, event sequences tend to be sparse and not equally spaced in time. As a result, they exhibit different properties, which are essential to account for when developing state-of-the-art methods. The community has paid little attention to the specifics of heterogeneous event sequences. Existing research in clustering primarily focuses on classic times series data. It is unclear if proposed methods in the literature generalize well to event sequences. Here we propose COHORTNEY as a novel deep learning method for clustering heterogeneous event sequences. Our contributions include (i) a novel method using a combination of LSTM and the EM algorithm and code implementation; (ii) a comparison of this method to previous research on time series and event sequence clustering; (iii) a performance benchmark of different approaches on a new dataset from the finance industry and fourteen additional datasets. Our results show that CO-HORTNEY vastly outperforms in speed and cluster quality the state-ofthe-art algorithm for clustering event sequences.

show abstract

“…Entries, such as [13], [7], [1], [19], validate the significance of topological data analysis (TDA) for time series data analysis. TDA is a set of statistical methods aimed to analyse complex datasets.…”

Section: Introductionmentioning

confidence: 99%

TOTOPO: Classifying univariate and multivariate time series with Topological Data Analysis

Pilyugina¹,

Rivera-Castro²,

Burnaev³

2020

Preprint

Self Cite

View full text Add to dashboard Cite

This work is devoted to a comprehensive analysis of topological data analysis for time series classification. Previous works have significant shortcomings, such as lack of large-scale benchmarking or missing state-of-the-art methods. In this work, we propose TOTOPO for extracting topological descriptors from different types of persistence diagrams. The results suggest that TOTOPO significantly outperforms existing baselines in terms of accuracy. TOTOPO is also competitive with the state-of-the-art, being the best on 20% of univariate and 40% of multivariate time series datasets. This work validates the hypothesis that TDA-based approaches are robust to small perturbations in data and are useful for cases where periodicity and shape help discriminate between classes.

show abstract

Topology-Based Clusterwise Regression for User Segmentation and Demand Forecasting

Cited by 5 publications

References 34 publications

The Net Promoter Score interrogated through a services marketing lens: Review and recommendations for service organizations

The Net Promoter Score interrogated through a services marketing lens: Review and recommendations for service organizations

COHORTNEY: Non-Parametric Clustering of Event Sequences

TOTOPO: Classifying univariate and multivariate time series with Topological Data Analysis

Contact Info

Product

Resources

About