The increasing trend of problem representation and high-dimensional data collection calls for the utilization of feature selection in many machines learning tasks and big data representations. However, identifying meaningful features from thousands of related features in the smart home data which are dissimilar in nature remains a nontrivial task. This has prompted for the deployment of a feature selection algorithm (FSA) that provides two possible solutions. First, to provide an efficient scheme that best optimizes the features for subsystem decisions and second, tackles feature subset selection bias problem. In this paper, a MFES framework for feature selection is proposed that uses a hybrid mechanism to tackle the problem of feature subset selection bias in intelligent building data. The mechanism uses the effectiveness of filters and accuracy of wrappers to obtain significant features for prediction. The proposed MFES framework resulted in 92.17

The entity of intelligent building is integrated with diversified service function of control, automation and communication of devices in its environment, and to perform them in joined manner via intelligent tasks. Rapid improvement in sensor technologies and advancement in electronics have given rise to heterogeneous systems growth in intelligent building. Most of these subsystems are dissimilar and not intended to perform interoperation task. Consequently, it is rather difficult to perform decision making with the combination of these systems considering the variety of data that are not efficient in adapting to the changing environment. One of the recent decision support solutions provided was Left–right Hidden Markov Model (LR-HMM) which uses left-right algorithm to improve accuracy of prediction based on single timely decision. However, it leads to low accuracy when multiple timely decisions are performed. Therefore, to ensure timely decision, the accuracy of prediction should be improved when performing multiple decisions. We propose a new decision model to improve performance in such situations. The goal is to improve the accuracy of prediction when multiple decisions are performed. Experiments are conducted to evaluate the performance of the proposed Re-estimated Ergodic Hidden Markov Model (RE-HMM), and show that it improves the average accuracy compared with LR-HMM. It is examined when tested on the Local Area Network (LAN) settings.

Query structuring systems are keyword search systems recently used for the effective retrieval of XML documents. Existing systems fail to put keyword query ambiguity prob-lems into consideration during query pre-processing and return irrelevant predicate nodes. As a result, these sys-tems return irrelevant results. In this research, an XML keyword search system, called N-gram based XML query structuring system (NBXQSS) is developed to improve the performance of keyword searches. The NBXQSS uses an N-gram Based Query Segmentation (NBQS) method which interprets a user query as a list of semantic units to help resolve ambiguity. The system also introduces an improved predicate identification algorithm (IPIA) to return rele-vant predicates. The IPIA uses a proposed function to com-pute the query term proximity and ordering. The effective-ness of the NBXQS is demonstrated through experimental performance study on some real-world XML documents. The results show that the developed system performs bet-ter compared to the existing system in terms of precision.