Customer Data Clustering via Weighted Preprocessing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current clustering algorithms for customer data in commercial databases are computationally complex and time-consuming, especially when determining the optimal number of clusters, leading to inefficiencies in identifying customer characteristics and potential market segments.
Innovation Solution
A method that preprocesses customer records by determining customer types, normalizing and weighting attribute values, and applying a clustering algorithm like K-means to efficiently divide records into pre-defined clusters, reducing computational complexity and adjusting dispersion and purity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If clustering algorithms are applied to customer data in commercial databases, then customer characteristics and market segments can be identified, but the computational complexity and time required for clustering increase significantly
Solution Approach 1:
The patent applies preliminary action by performing data preprocessing before clustering, including handling missing values, normalizing data, and selecting relevant features. This preliminary processing reduces the computational burden during the actual clustering phase and improves clustering efficiency without sacrificing accuracy
Solution Approach 2:
The patent changes parameters by transforming the original customer data into a normalized and weighted feature space. By adjusting parameter scales and weights during preprocessing, the algorithm achieves better clustering results with reduced computational complexity
2Measurement precision
If the number of clusters K is determined through trial with different values, then the optimal clustering result can be obtained, but the computational complexity increases
Solution Approach 1:
The patent implements feedback mechanisms through validation metrics that evaluate clustering quality. By using these feedback signals, the algorithm can determine the optimal K value more efficiently without requiring exhaustive trial of all possible K values, thus reducing computational complexity while maintaining result quality
3Productivity
If pre-processing steps are added to customer records, then clustering efficiency is improved, but the data processing steps increase
Solution Approach 1:
The patent segments the data processing task into distinct preprocessing modules (handling missing values, normalization, feature selection) and the main clustering algorithm. This segmentation allows each component to be optimized independently and improves overall productivity by making the preprocessing steps more manageable and efficient
Data Source
AI summary
A computer implemented method for clustering customers includes receiving a source set of customer records, wherein each customer record represents one customer, and each customer record includes at least one data attribute, and each data attribute has an attribute value; pre-processing the source set of customer records to generate a pre-processed set of customer records; executing a clustering algorithm on the pre-processed set of customer records to group the pre-processed set of customer records into clusters of a pre-defined number. The pre-processing comprises: determining the type of a customer in the source set of customer records; using a type attribute value to indicate the type of the customer in its customer record; normalizing data attribute values and type attribute values; weighting to the data attribute values and the type attribute values respectively to obtain weighted attribute values of the data attribute and weighted attribute values of the type attribute.