Dr H Shaheen

An efficient classifier decision tree for active context source discover on mobile pervasive environment

• Authors Details :  
• H.shaheen,  
• Dr.s.karthik

879 Views Research reports

Mobile pervasive environment interact with several devices at varying service ranges. The technical heterogeneity of pervasive environment is expected to increase the system flexibility and adaptability on modeling with context training phase. While working with context based training phase, time entity measure is considered as the significant issue. The evaluation of the services through numerous devices during training phase does not acquire an effective service monitoring on mobile pervasive environment. Mobile pervasive environment based information extraction fails to modify the patterns as activities change over time. To improve the flexibility of context training phase in mobile pervasive environment, an Active Context Source Discover Training Phase (ACSDTP) with Classifier Decision Tree Support (CDTS) mechanism is proposed in this paper. Our research work is to develop an effective modification (i.e., updation) of the pattern on training phase with real world context as per changes over time. Initially, the ACSDTP set up the available sensors in pervasive environment to work with the ever changing set of context users. The available sensors are maintained using the Active Discover process. Second, the CDTS mechanism is designed using weighted prediction for easy identification of context result on the training phase. Decision tree is operated separately using the learning techniques, where the identification is performed in a significant manner with minimal time factor. The learning process is performed to identify the inferred situations. Finally, the integration process is carried out to work with the complex association between the situations and sensor data in the mobile pervasive environment to achieve flexibility and adaptability factor. Experiment is conducted on factors such as time entity measure rate, precision ratio, and user context result determination level.

Article DOI & Crossmark Data

DOI : https://doi.org/10.37622/000000

Article Subject Details

• Allied Science
• Basic science
• Health sciences
• Information technology
• Interdisciplinary science
• Research
• Research methodology

Article File

Full Text PDF

---

---

---

More Article by Dr H Shaheen