Cloud service selection with multi-criteria decision making and artificial intelligence techniques

This thesis investigates using multi-criteria decision-making (MCDM) methods and artificial intelligence techniques in the cloud service selection process. Today, cloud computing plays a critical role in the digital transformation of businesses, and selecting the right cloud service provider is vital for the operational efficiency of organizations. However, numerous criteria and alternatives in cloud service selection make the decision-making process complex. This study aims to identify the most suitable cloud service provider. Through brainstorming with six experts in information technology and a literature review, eight key criteria were identified as important in selecting a cloud service provider: Cost, User-Friendly Interface and Management Tools, Data Backup and Recovery, Security, Integration, Support and Customer Services, Performance, and Compliance. This thesis employed the AHP (Analytic Hierarchy Process) and TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) methods from MCDM techniques in the cloud service selection process. AHP was used to determine the criteria weights, and TOPSIS was applied to select the most suitable cloud service provider. Machine learning methods for predicting cloud service providers were also compared. The k-Nearest Neighbors, Random Forest, Neural Network, and Gradient Boosting algorithms were selected for this comparison. According to the results, kNN was identified as the best prediction method based on the MAE, MAPE, and RMSE criteria. This approach provides a comprehensive framework for decision-makers to balance criteria and perform more accurate analyses on large datasets. While MCDM methods determine the importance of criteria and rank the alternatives, artificial intelligence techniques are used for prediction. In the application section of the study, the steps for data collection, the application of MCDM, and artificial intelligence methods for evaluating cloud service providers were discussed in detail. The results show that the proposed approach provides faster, more accurate, and effective outcomes in cloud service selection. In conclusion, this thesis offers a valuable methodology for utilizing MCDM and artificial intelligence techniques in complex decision-making processes such as cloud service selection. It highlights the advantages and limitations of this approach and provides recommendations for future research. The methodology aims to assist businesses in selecting cloud services more effectively, creating a more efficient and competitive digital environment.