Prediction of Bankruptcy among Middle 40 per Cent (M40) in Malaysia using Support Vector Machine
The Middle 40 Per Cent (M40) demographic comprises a substantial proportion of Malaysia's consumer base, comprising the median 40% of income earners. This group is a big part of the country's customer base. The rising cost of living, flat wage growth, and rising family debt make this group...
| id |
Sabri N.; Sahiq A.N.M.; Hamzah H.H.M.; Sain H.; Mangshor N.N.A.; Shari A.A. |
|---|---|
| spelling |
Sabri N.; Sahiq A.N.M.; Hamzah H.H.M.; Sain H.; Mangshor N.N.A.; Shari A.A. 2-s2.0-85219506277 Prediction of Bankruptcy among Middle 40 per Cent (M40) in Malaysia using Support Vector Machine 2024 2024 IEEE 22nd Student Conference on Research and Development, SCOReD 2024 10.1109/SCOReD64708.2024.10872749 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85219506277&doi=10.1109%2fSCOReD64708.2024.10872749&partnerID=40&md5=39999f86f8ddbf3849f9b2a7bc228357 The Middle 40 Per Cent (M40) demographic comprises a substantial proportion of Malaysia's consumer base, comprising the median 40% of income earners. This group is a big part of the country's customer base. The rising cost of living, flat wage growth, and rising family debt make this group especially vulnerable to financial instability. Given these economic pressures, predicting the possibility of bankruptcy in this population is becoming increasingly crucial to allow for early intervention and preventive actions. Support Vector Machines (SVM) are recognized for their efficacy in classification tasks, particularly with small or imbalanced datasets. SVM prediction will be utilized using 14694 samples of the Debt Management Program (DMP) dataset from 2016 to 2020. Twenty-three (23) features will be implemented on the dataset and Synthetic Minority Oversampling Technique (SMOTE) to improve the imbalanced datasets. The experiment demonstrates that the SVM classifier obtained 91.1% accuracy when implementing PUK kernels. It suggests that SVM is a plausible alternative model for predicting personal bankruptcy in the M40 group. To anticipate personal bankruptcy for M40, additional prediction models will be evaluated in the future. © 2024 IEEE. Institute of Electrical and Electronics Engineers Inc. English Conference paper |
| author |
2-s2.0-85219506277 |
| spellingShingle |
2-s2.0-85219506277 Prediction of Bankruptcy among Middle 40 per Cent (M40) in Malaysia using Support Vector Machine |
| author_facet |
2-s2.0-85219506277 |
| author_sort |
2-s2.0-85219506277 |
| title |
Prediction of Bankruptcy among Middle 40 per Cent (M40) in Malaysia using Support Vector Machine |
| title_short |
Prediction of Bankruptcy among Middle 40 per Cent (M40) in Malaysia using Support Vector Machine |
| title_full |
Prediction of Bankruptcy among Middle 40 per Cent (M40) in Malaysia using Support Vector Machine |
| title_fullStr |
Prediction of Bankruptcy among Middle 40 per Cent (M40) in Malaysia using Support Vector Machine |
| title_full_unstemmed |
Prediction of Bankruptcy among Middle 40 per Cent (M40) in Malaysia using Support Vector Machine |
| title_sort |
Prediction of Bankruptcy among Middle 40 per Cent (M40) in Malaysia using Support Vector Machine |
| publishDate |
2024 |
| container_title |
2024 IEEE 22nd Student Conference on Research and Development, SCOReD 2024 |
| container_volume |
|
| container_issue |
|
| doi_str_mv |
10.1109/SCOReD64708.2024.10872749 |
| url |
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85219506277&doi=10.1109%2fSCOReD64708.2024.10872749&partnerID=40&md5=39999f86f8ddbf3849f9b2a7bc228357 |
| description |
The Middle 40 Per Cent (M40) demographic comprises a substantial proportion of Malaysia's consumer base, comprising the median 40% of income earners. This group is a big part of the country's customer base. The rising cost of living, flat wage growth, and rising family debt make this group especially vulnerable to financial instability. Given these economic pressures, predicting the possibility of bankruptcy in this population is becoming increasingly crucial to allow for early intervention and preventive actions. Support Vector Machines (SVM) are recognized for their efficacy in classification tasks, particularly with small or imbalanced datasets. SVM prediction will be utilized using 14694 samples of the Debt Management Program (DMP) dataset from 2016 to 2020. Twenty-three (23) features will be implemented on the dataset and Synthetic Minority Oversampling Technique (SMOTE) to improve the imbalanced datasets. The experiment demonstrates that the SVM classifier obtained 91.1% accuracy when implementing PUK kernels. It suggests that SVM is a plausible alternative model for predicting personal bankruptcy in the M40 group. To anticipate personal bankruptcy for M40, additional prediction models will be evaluated in the future. © 2024 IEEE. |
| publisher |
Institute of Electrical and Electronics Engineers Inc. |
| issn |
|
| language |
English |
| format |
Conference paper |
| accesstype |
|
| record_format |
scopus |
| collection |
Scopus |
| _version_ |
1828987861518516224 |