Comparison of distance metrics for hierarchical data in medical databases
Distance metrics are broadly used in different research areas and applications, such as bio-informatics, data mining and many other fields. However, there are some metrics, like pg-gram and Edit Distance used specifically for data with a hierarchical structure. Other metrics used for non-hierarchica...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Published: |
2014
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/3349/ |
| _version_ | 1848791006742315008 |
|---|---|
| author | Hassan, Diman Aickelin, Uwe Wagner, Christian |
| author_facet | Hassan, Diman Aickelin, Uwe Wagner, Christian |
| author_sort | Hassan, Diman |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | Distance metrics are broadly used in different research areas and applications, such as bio-informatics, data mining and many other fields. However, there are some metrics, like pg-gram and Edit Distance used specifically for data with a hierarchical structure. Other metrics used for non-hierarchical data are the geometric and Hamming metrics. We have applied these metrics to The Health Improvement Network (THIN) database which has some hierarchical data. The THIN data has to be converted into a tree-like structure for the first group of metrics. For the second group of metrics, the data are converted into a frequency table or matrix, then for all metrics, all distances are found and normalised. Based on this particular data set, our research question: which of these metrics is useful for THIN data?. This paper compares the metrics, particularly the pogram metric on finding the similarities of patients' data. It also investigates the similar patients who have the same close distances as well as the metrics suitability for clustering the whole patient population. Our results show that the two groups of metrics perform differently as they represent different structures of the data. Nevertheless, all the metrics could represent some similar data of patients as well as discriminate sufficiently well in clustering the patient population using k-means clustering algorithm. |
| first_indexed | 2025-11-14T18:21:39Z |
| format | Conference or Workshop Item |
| id | nottingham-3349 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| last_indexed | 2025-11-14T18:21:39Z |
| publishDate | 2014 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-33492020-05-04T16:54:35Z https://eprints.nottingham.ac.uk/3349/ Comparison of distance metrics for hierarchical data in medical databases Hassan, Diman Aickelin, Uwe Wagner, Christian Distance metrics are broadly used in different research areas and applications, such as bio-informatics, data mining and many other fields. However, there are some metrics, like pg-gram and Edit Distance used specifically for data with a hierarchical structure. Other metrics used for non-hierarchical data are the geometric and Hamming metrics. We have applied these metrics to The Health Improvement Network (THIN) database which has some hierarchical data. The THIN data has to be converted into a tree-like structure for the first group of metrics. For the second group of metrics, the data are converted into a frequency table or matrix, then for all metrics, all distances are found and normalised. Based on this particular data set, our research question: which of these metrics is useful for THIN data?. This paper compares the metrics, particularly the pogram metric on finding the similarities of patients' data. It also investigates the similar patients who have the same close distances as well as the metrics suitability for clustering the whole patient population. Our results show that the two groups of metrics perform differently as they represent different structures of the data. Nevertheless, all the metrics could represent some similar data of patients as well as discriminate sufficiently well in clustering the patient population using k-means clustering algorithm. 2014-09-04 Conference or Workshop Item PeerReviewed Hassan, Diman, Aickelin, Uwe and Wagner, Christian (2014) Comparison of distance metrics for hierarchical data in medical databases. In: Proceedings of the 2014 World Congress on Computational Intelligence (WCCI 2014), 6-11 July 2014, Beijing, China. Biomedical Informatics Data Mining Machine Learning http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6889554 |
| spellingShingle | Biomedical Informatics Data Mining Machine Learning Hassan, Diman Aickelin, Uwe Wagner, Christian Comparison of distance metrics for hierarchical data in medical databases |
| title | Comparison of distance metrics for hierarchical data in medical databases |
| title_full | Comparison of distance metrics for hierarchical data in medical databases |
| title_fullStr | Comparison of distance metrics for hierarchical data in medical databases |
| title_full_unstemmed | Comparison of distance metrics for hierarchical data in medical databases |
| title_short | Comparison of distance metrics for hierarchical data in medical databases |
| title_sort | comparison of distance metrics for hierarchical data in medical databases |
| topic | Biomedical Informatics Data Mining Machine Learning |
| url | https://eprints.nottingham.ac.uk/3349/ https://eprints.nottingham.ac.uk/3349/ |