INFERENSI PROBABILISTIK HYBRID IT VALUE MODEL MENGGUNAKAN BAYESIAN NETWORK

• Main Author: Ramadhani - NIM: 23515011 , Imaniar
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/28008
• Institution: Institut Teknologi Bandung

Nowadays, the changing of IT perspective causes a lot of research to measure the influence of IT values in organizations. Measuring the effect of hybrid IT value model is a model that has been proven to better explain the relationship between variables. In addition, the calculation method using a probabilistic approach will be used in this study to find out how much the influence of causal relationships between variables on the IT value model. <br /> <br /> <br /> <br /> <br /> The magnitude of the influence of causal relationships between variables is one form of uncertainty in the value of influence. The effect of causal relationships can be expressed in probabilistic form using the Bayesian Network method. This method has been proven to be used to integrate various knowledge and conditions has flexibly for the purpose of inference and diagnosis. Bayesian Network presents unlimited input data source variances and each variable of IT value can be identified. <br /> <br /> <br /> <br /> <br /> Calculation of causal relationships between variables in a hybrid IT value model using a probabilistic approach includes the structure, nature, and direction of the relationships between variables. This is an important part in ensuring confidence in the influence of IT values in the organization. The guarantee of this belief is expected to be a solution to maintain the accuracy of the influence of IT values. <br /> <br /> <br /> <br /> <br /> The purpose of this study is to obtain probabilistic inference based on hybrid IT value models using Bayesian Network to determine causal relationships between variables. This will be explained in the probability equation model and the graphical model. The results of the study show that the probabilistic approach with Bayesian Network can measure how much the confidence of relationships between variables. In addition, this approach can refer and diagnose relationships between variables based on graphic models in the best and worst conditions. As a result, adaptability and model flexibility can be more measurable. <br />