The amount of data generated, collected, and processed through computer networks has increased exponentially in recent years. Network attacks have also become an inherent concern in complex networks as a result of this increase of data. The practise of assessing trust using attributes that influence trust is known as trust evaluation. It is confronted with a number of serious challenges, including a shortage of critical assessment data, a requirement for big data processing, a call for a simple trust relationship expression, and the expectation of automation. Machine learning (ML) has been applied to trust evaluation in order to overcome these issues and intelligently and automatically evaluate trust. This research propose novel technique in data fusion model with security and data optimization technique in edge computing. Here the proposed data fusion is carried out using secure sequential discriminant auto encoder in which the improvement of data accuracy, as well as for the maximizing of Edge-cloud based sensor networks lifespan. The fusion of edge cloud data has been carried out using discriminant auto encoder which is integrated with distributed edge cloud users, where the security of the network has been enhanced using secure sequential fuzzy based trust model. The data optimization has been established using Genetic swarm lightweight optimization algorithm. The experimental analysis has been carried out based on data fusion as well as network security model. The parametric analysis is carried out in terms of network security analysis, throughput, Coverage fraction, Delay time, Energy consumption, Storage efficiency. © 2022 Elsevier B.V., All rights reserved.