No abstract

Swarm intelligence systems are made up of a population of simple agents interacting locally with each another and with their environment. Artificial bee colony (ABC) algorithm, particle swarm optimization (PSO), ant colony optimization (ACO), differential evolution (DE) etc, are some example of swarm intelligence. In this work, an efficient modified version of ABC algorithm is proposed, where two additional operator crossover and mutation operator is used in the ABC algorithm. Here Crossover operator is used after the employed bee phase and mutation operator is used after scout bee phase of ABC algorithm. Proposed algorithm is applied at standard travelling salesman problem (TSP) for checking the efficiency of proposed algorithm and also simulated results are compared with ABC with uniform mutation algorithm and Basic ABC algorithm. The simulated result showed that the proposed algorithm is better than all the modified version of ABC algorithm.

In this paper, the view variations effect in human gait recognition using sub-window extraction algorithm is proposed. Here different variation is created based on the walking people in three different angles (i.e. 0 0 , 45 0 and 90 0 )with respect to particular line. Our proposed method works on two different phases: Extraction phase and Recognition Phase. In first phase, gait images, captured from different angles, are enhanced using clipping, filtering and histogram equalization.Then apply proposed sub window extraction algorithm on enhanced gait images and gathered different features like person length, leg angle, leg length, hand length etc. Finally apply back propagation algorithm for the recognition of gait images. Experiments are carried out using different datasets.

No abstract

For a successful project development, it is important for any software organization that the project should be completed within time and budget, and the project should have requisite quality. This paper presents an Ensemble learning based Adaptive Neuro-Fuzzy Approach for Software Development Time Estimation. The concept behind this technique is based on ensemble learning methods. This technique combines multiple models into one model. The ensemble fits a new learner to the difference between the experiential response and the aggregated prediction of all learners which grown previously. In this paper, we describe a brief literature review of different techniques of software development time estimation along with a comparison of different techniques with our approach.