Muhammad, Faisal (2024) Implementation of Hybrid Indexing, Clustering and Classification Methods to Enhance Rural Development Programme in South Sulawesi. Doctoral thesis, Asia e University.
![[img]](http://ur.aeu.edu.my/style/images/fileicons/text.png) |
Text
Thesis Muhammad Faisal.pdf Restricted to Repository staff only Download (4MB) |
|
Text
Thesis Muhammad Faisal-1-24.pdf Download (346kB) |
Abstract
Limited involvement of communities and village institutions in planning, implementing, and supervising village development activities, as well as difficulty in monitoring development results, resulted in a lot of untapped natural potential in traditionally managed villages due to technological limitations and low levels of education. Development in rural areas faces obstacles from internal factors such as lack of initiative and knowledge among rural communities and external factors such as constraints on government support. Village development and assistance program is a strategy implemented to accelerate socio cultural development to increase the capacity of village governance and more organized administration. The village assistance programme can motivate and engage rural communities during the participatory and transparent phase of village development. The criteria used in this study were produced through the identification, verification, and validation stages by experts consisting of academian, government, and researchers. The objectives of this research are to develop a community standard of living index based on verified criteria collected from selected communities, to cluster the village based on the community standard of living index, to classify a village based on the Developing Village Index(DVI), Human Development Index(HDI), and Community Standard of Living Index(CSLI), and to map the relevant experts with the priority needs of a village based on the input from the villages. The initial stage in this study involved the design of the questionnaire, the process of criteria weighting, and the scoring of villages by communities. CSLI was developed to represent the community welfare level for each village. Clustering techniques such as Self-Organizing Map, Fuzzy C-Means, and Xie-Beny methods are utilized to clustering villages according to the Community Standard of Living Index. The Fuzzy Tsukamoto and Smallest of Maximum methods were then used to classify villages into less development, which involved CSLI-Clusters as indicators. Using the cosine similarity algorithm for knowledge recommendation is village identified, utilizing community feedback as the foundation. Based on the clustering results using CSLI Score, Head of Family, and Number of Residence criteria, it is stated that all villages are divided into 3 clusters, which are CSLI-Good, CSLI-Average, and CSLI-Poor. The classification technique using the CSLI-Cluster, DVI, and HDI criteria showed that as many as 22 villages had the status of Less Development level, and 8 villages were declared Developed. This research identified the following recommended fields: agricultural science in 11 villages, social sciences in 11 villages, economics in 10 villages, entrepreneurship in 8 villages, marine science,forestry, and computers in 11 villages each, and regional planning in 2 villages with 82% accuracy. The result validation of decisions on the placement of accompanying experts in each village with actual data was carried out using the confusion matrix metho, which are accurate = 0.819, precision = 1, recall = 0.819, and F1 Score = 0.9. This shows that the accuracy status indicates a high percentage of correct predictions, and then the F1 Score of 0.9 indicates a well-balanced trade-off between precision and recall, demonstrating the model's overall effectiveness. The government can use the findings of this research as a decision-making tool regarding equitable village development programs.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Uncontrolled Keywords: | Rural, village assistant, indexing, clustering, classification, recommendation, DSS |
| Divisions: | School of Graduate Studies |
| Depositing User: | Siti Nor Fairuz Rosaidee |
| Date Deposited: | 02 Jan 2025 08:07 |
| Last Modified: | 02 Jan 2025 08:37 |
| URI: | http://ur.aeu.edu.my/id/eprint/1264 |
Actions (login required)
 |
View Item |
