Papers by Kothalil Gopalakrishnan Anilkumar
Neural Network Based Priority Assignment for Job Scheduler
This paper describes the design and implementation of a neural network-based job priority assigne... more This paper describes the design and implementation of a neural network-based job priority assigner system for a job scheduling environment. Scheduling deals with the allocation of resources over time to perform a collection of tasks. Scheduling problems arise in domains as diverse as manufacturing, computer processing, transportation, health care, space exploration, and education. In the case of a neural network (NN) based scheduler, once the job attributes are properly trained for a specified schedule, it will never miss that related scheduling pattern for that particular job. An NN based scheduling procedure can successfully overcome the local minima of its error surface. This paper reports on research which established that a back propagation neural network-based priority procedure would recognize jobs from a job queue by estimating each job’s priority. Once the priorities are assigned, it is not possible to alter the priorities under any circumstances.
A Thai word pronunciation simulator based on DFT analysis and GA search
The 2013 10th International Joint Conference on Computer Science and Software Engineering, May 29, 2013
ABSTRACT This paper aims to present and discuss the concept of a Thai word pronunciation simulato... more ABSTRACT This paper aims to present and discuss the concept of a Thai word pronunciation simulator based on DFT (Discrete Fourier Transform) analysis and GA (Genetic Algorithm) search. By DFT analysis, the simulator converts the sound of a Thai word from a foreign leaner into its frames to get sound segments. The GA search on the sound segments generates a highest similarity sound value with a standard Thai sound. The GA search is supported by a library of original Thai word sounds. Once the simulator generates a real pronunciation of the leaner’s word, it will store the sound data to train the user. The simulations presented in this paper indicate that the proposed approach is one of the most effective strategies of structuring a language trainer.
Neural Network Based Generalized Job-Shop Scheduler
Abstract. This paper presents a feed-forward neural network (NN), together with an alignment algo... more Abstract. This paper presents a feed-forward neural network (NN), together with an alignment algorithm to solve a generalized job-shop scheduling problem, one of NP-complete constraint satisfaction problems. The proposed NN is embedded with predefined criteria which is ...
Generalized job-shop scheduler using feed forward neural network and greedy alignment procedure
Artificial Intelligence and Applications, 2007
GENERALIZED JOB-SHOP SCHEDULER USING FEED FORWARD NEURAL NETWORK AND GREEDY ALIGNMENT PROCEDURE K... more GENERALIZED JOB-SHOP SCHEDULER USING FEED FORWARD NEURAL NETWORK AND GREEDY ALIGNMENT PROCEDURE Kothalil Gopalakrishnan Anilkumar and Thitipong Tanprasert Department of Computer Science, Faculty of Science and Technology ...
This paper presents a subjective job scheduler based on a 3-layer Backpropagation Neural Network ... more This paper presents a subjective job scheduler based on a 3-layer Backpropagation Neural Network (BPNN) and a greedy alignment procedure in order formulates a real-life situation. The BPNN estimates critical values of jobs based on the given subjective criteria. The scheduler is formulated in such a way that, at each time period, the most critical job is selected from the job queue and is transferred into a single machine before the next periodic job arrives. If the selected job is one of the oldest jobs in the queue and its deadline is less than that of the arrival time of the current job, then there is an update of the deadline of the job is assigned in order to prevent the critical job from its elimination. The proposed satisfiability criteria indicates that the satisfaction of the scheduler with respect to performance of the BPNN, validity of the jobs and the feasibility of the scheduler.
Generalized job-shop scheduler using feed forward neural network and greedy alignment procedure
GENERALIZED JOB-SHOP SCHEDULER USING FEED FORWARD NEURAL NETWORK AND GREEDY ALIGNMENT PROCEDURE K... more GENERALIZED JOB-SHOP SCHEDULER USING FEED FORWARD NEURAL NETWORK AND GREEDY ALIGNMENT PROCEDURE Kothalil Gopalakrishnan Anilkumar and Thitipong Tanprasert Department of Computer Science, Faculty of Science and Technology ...
This paper presents a feed-forward neural network (NN), together with an alignment algorithm to s... more This paper presents a feed-forward neural network (NN), together with an alignment algorithm to solve a generalized job-shop scheduling problem, one of NP-complete constraint satisfaction problems. The proposed NN is embedded with predefined criteria which is relevant to determine starting time of various operations within a job and is trained until its mean squared error is reduced to a value less than 10%. The output of the NN is used by the alignment algorithm to detect the precedence order of each operation within a job and schedule each operation on respective machine without violating the detected precedence order (precedence constraint) and resource constraint. The key findings reveal that the NN plays a decisive role to estimate the starting time of each operation within a job before the operations are scheduled on a respective machine. Simulations of the proposed scheduler have shown that the NN with the alignment algorithm approach is efficient with respect to the quality of expected solutions and the solving speed.
This paper describes the design and implementation of a neural network-based job priority assigne... more This paper describes the design and implementation of a neural network-based job priority assigner system for a job scheduling environment. Scheduling deals with the allocation of resources over time to perform a collection of tasks. Scheduling problems arise in domains as diverse as manufacturing, computer processing, transportation, health care, space exploration, and education. In the case of a neural network (NN) based scheduler, once the job attributes are properly trained for a specified schedule, it will never miss that related scheduling pattern for that particular job. An NN based scheduling procedure can successfully overcome the local minima of its error surface. This paper reports on research which established that a back propagation neural network-based priority procedure would recognize jobs from a job queue by estimating each job's priority. Once the priorities are assigned, it is not possible to alter the priorities under any circumstances.
Uploads
Papers by Kothalil Gopalakrishnan Anilkumar