MC-SVM Based Work Flow Preparation in Cloud with Named Entity Identification

Abstract

The existing applications that are associated with the internet produce enormous amount of data according to the requirements of diverse circumstances prevailing. This causes multitudes of challenges in examining the data and as well as in the operation of the system that relies on the cloud. To simply process and manage the execution of the tasks properly with respect to time the workflow scheduling was devised in the cloud. To further enhance the process of scheduling the named entity recognition is used. The NER-named entity recognition is an important chore of more general discipline of internet explorer application. Since the NER- problem is highly challenging in cloud paradigm. An innovative frame work termed as the MC-SVM (Multi Class- Support Vector Machine) is laid out in the paper to devise the scheduling of the workflow in the cloud paradigm. The scheduling of the tasks in the cloud delivers an arrangement setting up the work flows with the named entity recognition using the MC-SVM. The algorithm developed enhances the resource allocation process, by performing a simultaneous and dynamic allocation/reallocation of named entities to the resources of the cloud satisfying the demands in the performance and cost. The results observed on validating the proposed algorithm proves the capability of the system to manage the resources in the cloud effectively optimizing the make span and the cost.