Detecting Multi-Relationship Links in Sparse Datasets

Nie, Dongyun; Roantree, Mark

doi:10.5220/0007696901490157

Cited by 4 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The net result is that companies may not be aware that two different policies belong to the same customer. In previous work (Nie and Roantree, 2019), we addressed the linkage problem but this was merely Step 1 in our overall approach. This section of the paper focuses on the extraction of a principal data set and using these variables as input to the customer classification process.…”

Section: Methodsmentioning

confidence: 99%

“…The ultimate goal of this research is to provide a CLV ranking for each customer in our collaborator's data set and in previous research Nie and Roantree (2019), we performed the record linkage process to deliver unified customer records. However, many of the parameters required for CLV calculations are missing from this data set and the final process will require a level of imputation to generate those missing values.…”

Section: Problem Description and Motivationmentioning

confidence: 99%

“…However, there are very high numbers of individuals who cannot be matched as common identifiers do not exist across data sets. This step is not part of this research but was presented in Nie and Roantree (2019). In this earlier work, a customer matching approach was developed using a modified form of agglomerative hierarchical clustering (AHC), together with a companion ruleset to detect and link components of the same customer record and also clients with family members.…”

Section: Record Linkagementioning

confidence: 99%

See 2 more Smart Citations

A methodology for classification and validation of customer datasets

Nie

Cappellari

Roantree

2020

JBIM

Self Cite

View full text Add to dashboard Cite

Purpose The purpose of this paper is to develop a method to classify customers according to their value to an organization. This process is complicated by the disconnected nature of a customer record in an industry such as insurance. With large numbers of customers, it is of significant benefit to managers and company analysts to create a broad classification for all customers. Design/methodology/approach The initial step is to construct a full customer history and extract a feature set suited to customer lifetime value calculations. This feature set must then be validated to determine its ability to classify customers in broad terms. Findings The method successfully classifies customer data sets with an accuracy of 90%. This study also discovered that by examining the average value for key variables in each customer segment, an algorithm can label the group of clusters with an accuracy of 99.3%. Research limitations/implications Working with a real-world data set, it is always the case that some features are unavailable as they were never recorded. This can impair the algorithm’s ability to make good classifications in all cases. Originality/value This study believes that this research makes a novel contribution as it automates the classification of customers but in addition, the approach provides a high-level classification result (recall and precision identify the best cluster configuration) and detailed insights into how each customer is classified by two validation metrics. This supports managers in terms of market spend on new and existing customers.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Problem Description and Motivationmentioning

confidence: 99%

Section: Record Linkagementioning

confidence: 99%

See 1 more Smart Citation

A methodology for classification and validation of customer datasets

Nie

Cappellari

Roantree

2020

JBIM

Self Cite

View full text Add to dashboard Cite

show abstract

Predicting Customer Churn for Insurance Data

Scriney

Nie

Roantree

2020

Big Data Analytics and Knowledge Discovery

View full text Add to dashboard Cite

An Ostensive Information Architecture to Enhance Semantic Interoperability for Healthcare Information Systems

2023

View full text Add to dashboard Cite

Semantic interoperability establishes intercommunications and enables data sharing across disparate systems. In this study, we propose an ostensive information architecture for healthcare information systems to decrease ambiguity caused by using signs in different contexts for different purposes. The ostensive information architecture adopts a consensus-based approach initiated from the perspective of information systems re-design and can be applied to other domains where information exchange is required between heterogeneous systems. Driven by the issues in FHIR (Fast Health Interoperability Resources) implementation, an ostensive approach that supplements the current lexical approach in semantic exchange is proposed. A Semantic Engine with an FHIR knowledge graph as the core is constructed using Neo4j to provide semantic interpretation and examples. The MIMIC III (Medical Information Mart for Intensive Care) datasets and diabetes datasets have been employed to demonstrate the effectiveness of the proposed information architecture. We further discuss the benefits of the separation of semantic interpretation and data storage from the perspective of information system design, and the semantic reasoning towards patient-centric care underpinned by the Semantic Engine.

show abstract

Detecting Multi-Relationship Links in Sparse Datasets

Cited by 4 publications

References 0 publications

A methodology for classification and validation of customer datasets

A methodology for classification and validation of customer datasets

Predicting Customer Churn for Insurance Data

An Ostensive Information Architecture to Enhance Semantic Interoperability for Healthcare Information Systems

Contact Info

Product

Resources

About