Feature Selection with Graph Mining Technology
Many real world applications have problems with high dimensionality, which existing algorithms cannot overcome. A critical data preprocessing problem is feature selection, whereby its non-scalability negatively influences both the efficiency and performance of big data applications. In this research...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tsinghua University Press
2019-06-01
|
| Series: | Big Data Mining and Analytics |
| Subjects: | |
| Online Access: | https://www.sciopen.com/article/10.26599/BDMA.2018.9020032 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832572804679073792 |
|---|---|
| author | Thosini Bamunu Mudiyanselage Yanqing Zhang |
| author_facet | Thosini Bamunu Mudiyanselage Yanqing Zhang |
| author_sort | Thosini Bamunu Mudiyanselage |
| collection | DOAJ |
| description | Many real world applications have problems with high dimensionality, which existing algorithms cannot overcome. A critical data preprocessing problem is feature selection, whereby its non-scalability negatively influences both the efficiency and performance of big data applications. In this research, we developed a new algorithm to reduce the dimensionality of a problem using graph-based analysis, which retains the physical meaning of the original high-dimensional feature space. Most existing feature-selection methods are based on a strong assumption that features are independent of each other. However, if the feature-selection algorithm does not take into consideration the interdependencies of the feature space, the selected data fail to correctly represent the original data. We developed a new feature-selection method to address this challenge. Our aim in this research was to examine the dependencies between features and select the optimal feature set with respect to the original data structure. Another important factor in our proposed method is that it can perform even in the absence of class labels. This is a more difficult problem that many feature-selection algorithms fail to address. In this case, they only use wrapper techniques that require a learning algorithm to select features. It is important to note that our experimental results indicates, this proposed simple ranking method performs better than other methods, independent of any particular learning algorithm used. |
| format | Article |
| id | doaj-art-51216d57c7ae4f39aa729063d30bc32f |
| institution | Kabale University |
| issn | 2096-0654 |
| language | English |
| publishDate | 2019-06-01 |
| publisher | Tsinghua University Press |
| record_format | Article |
| series | Big Data Mining and Analytics |
| spelling | doaj-art-51216d57c7ae4f39aa729063d30bc32f2025-02-02T06:50:33ZengTsinghua University PressBig Data Mining and Analytics2096-06542019-06-0122738210.26599/BDMA.2018.9020032Feature Selection with Graph Mining TechnologyThosini Bamunu Mudiyanselage0Yanqing Zhang1<institution content-type="dept">Department of Computer Science</institution>, <institution>Georgia State University</institution>, <city>Atlanta</city>, <state>GA</state> <postal-code>30302</postal-code>, <country>USA</country>.<institution content-type="dept">Department of Computer Science</institution>, <institution>Georgia State University</institution>, <city>Atlanta</city>, <state>GA</state> <postal-code>30302</postal-code>, <country>USA</country>.Many real world applications have problems with high dimensionality, which existing algorithms cannot overcome. A critical data preprocessing problem is feature selection, whereby its non-scalability negatively influences both the efficiency and performance of big data applications. In this research, we developed a new algorithm to reduce the dimensionality of a problem using graph-based analysis, which retains the physical meaning of the original high-dimensional feature space. Most existing feature-selection methods are based on a strong assumption that features are independent of each other. However, if the feature-selection algorithm does not take into consideration the interdependencies of the feature space, the selected data fail to correctly represent the original data. We developed a new feature-selection method to address this challenge. Our aim in this research was to examine the dependencies between features and select the optimal feature set with respect to the original data structure. Another important factor in our proposed method is that it can perform even in the absence of class labels. This is a more difficult problem that many feature-selection algorithms fail to address. In this case, they only use wrapper techniques that require a learning algorithm to select features. It is important to note that our experimental results indicates, this proposed simple ranking method performs better than other methods, independent of any particular learning algorithm used.https://www.sciopen.com/article/10.26599/BDMA.2018.9020032graph miningnetwork embeddingbig data analysisfeature selectionhigh-dimensional data |
| spellingShingle | Thosini Bamunu Mudiyanselage Yanqing Zhang Feature Selection with Graph Mining Technology Big Data Mining and Analytics graph mining network embedding big data analysis feature selection high-dimensional data |
| title | Feature Selection with Graph Mining Technology |
| title_full | Feature Selection with Graph Mining Technology |
| title_fullStr | Feature Selection with Graph Mining Technology |
| title_full_unstemmed | Feature Selection with Graph Mining Technology |
| title_short | Feature Selection with Graph Mining Technology |
| title_sort | feature selection with graph mining technology |
| topic | graph mining network embedding big data analysis feature selection high-dimensional data |
| url | https://www.sciopen.com/article/10.26599/BDMA.2018.9020032 |
| work_keys_str_mv | AT thosinibamunumudiyanselage featureselectionwithgraphminingtechnology AT yanqingzhang featureselectionwithgraphminingtechnology |