Arun Malik ¹, Babita Pandey ²

Affiliations
1 Department of Computer Science and Engineering, Lovely Professional University, Jalandhar - 144411, Punjab, IN
2 Department of Computer Applications, Lovely Professional University, Jalandhar - 144411, Punjab, IN

Abstract

Objectives: The main objective of this paper is to find the best method for collecting the data for VANET in terms of communication overhead, average latency and packet delivery ratio. Methods/Analysis: OMNet++ is used to compare the existing techniques for data collection in VANET at single RSU. Performance Index is calculated by finding the Communication Overhead (CO), average latency and Packet Delivery Ratio (PDR) for each scheme of data collection. The scheme with higher value of performance index will outperform others. Simulation parameters like dimension of space, minimum velocity and maximum velocity of vehicles are kept constant. Findings: Comparison of different Data Collection Schemes is done on the basis of performance index. Performance Index shows the effectiveness of data collection schemes. PI is based on various parameters like CO, Latency, and PDR. PI of a data collection scheme decreases when CO increases and vice-versa. So, if number of messages will increase CO will increase and hence will decrease the PI. PI of a data collection scheme decreases when latency increases and vice-versa. So, if time delay is more latency will increase and hence will decrease the PI. PI of a Data Collection Schemes increases when PDR increases and vice-versa. So, if number of packets received by a destination will be more PDR will increase and hence will increase the PI. Analysis of results show that PI of VIB-CP is the best among all DCSs, as it has less CO, low latency, and high PDR. Novelty/Improvement: Comparison of heterogeneous Data Collection Schemes is made on same parameters and their performance index is measured. Out of all existing schemes it is analyzed that VIB-CP outperforms.

Keywords

Communication Overhead (CO), Data Collection Schemes (DCS), Data Mining (DM), Packet Delivery Ratio (PDR), Performance Index (PI)

Full Text

     

Aditya Khamparia ¹, Babita Pandey ²

Affiliations
1 Department of Computer Science and Engineering, Lovely Professional University, Phagwara – 144411, Punjab, IN
2 Department of Computer Applications, Lovely Professional University, Phagwara – 144411, Punjab, IN

Abstract

Objectives: The major goal of this research paper is to present a QoS and cognitive parameter based model for selection of semantic web services. The presented model provides a completely novel and formalized measurement of different cognitive parameters. Methods/Statistical analysis: Rule based model is used for describing hierarchical relationships among QoS and cognitive parameters. The short life factor is used for dealing with known certainties lies in these parameters. The certainty factor is computed by using a measure of belief and measure of disbelief. Finally, the computed result is based on the satisfaction level of consumer agent. Findings: The rule base model generated from the hierarchal structure is used for computing CCF of each qualitative and quantitative parameter. As the rule base is generated from the hierarchical tree therefore as tree changes the rule base also changes. It is observed from the result that the overall computational overhead is very less in this cognitive based uncertainty model; it leads to fast, efficient and smart retrieval or selection of services for consumer agent. The proposed approach overcomes limitations of different models by combining several cognitive parameters, focusing on user’s preferences on QoS attributes in an efficient way. Application/Improvements: The predicted applications of proposed model in E-learning, E-governance based systems and identification of web services. The generated rule base is large so by adapting neuro symbolic rules the rule base could be reduced to provide efficient and fast delivery of services.

Keywords

Certainty Factor, Cognitive Parameters, QoS, Rule Based, Short Life.

Full Text

   

Divya ¹, Babita Pandey ², Devendra K. Pandey ³

Affiliations
1 School of Computer Science and Engineering, Lovely Professional University, Chaheru, Phagwara - 144411, Punjab, IN
2 School of Computer Applications, Lovely Professional University, Chaheru, Phagwara - 144411, Punjab, IN
3 School of Biosciences, Lovely Professional University, Chaheru, Phagwara - 144411, Punjab, IN

Abstract

Background/Objectives: Microarray technology allows the neuromuscular dystrophy to be predicted using gene expression patterns. Microarray gene expression data suffer from curse of high dimensionality i.e. tens of thousands of genes and few samples. So, it is necessitate reducing the dimension for accurate diagnosis. Methods/Statistical Analysis: Firstly, five-fold cross validation technique is applied to generate random results. Two feature selection techniques i.e. t-test and entropy are employed to select the genes. K-nearest neighbor and linear support vector machine are deployed for classification of diseased samples with the help of ranked genes. The performance of these integrated techniques is tested on the microarray dataset of neuromuscular dystrophies i.e. Juvenile Dermatomyositis (JDM) and Fascioscapulohumeral Muscular Dystrophy (FSHD). Findings: Effective disease specific genes are selected from thousand of genes. The value of various performance measures shows that the integration of entropy with k-nearest neighbor has outperformed on both datasets. It has given 89.47% accuracy on JDM dataset and 100% accuracy on FSHD dataset. The integration of these methods is first time application on these two diseases datasets. It can be applied on other neuromuscular disorder datasets as well.

Keywords

Dimension Reduction, Entropy, K-Fold Validation, K-Nearest Neighbor, Neuromuscular Dystrophy, Support Vector Machine.

Full Text

   

Arun Malik ¹, Babita Pandey ²

Affiliations
1 Department of Computer Science and Engineering, Lovely Professional University, Jalandhar - 144411, Punjab, IN
2 Department of Computer Applications, Lovely Professional University, Jalandhar - 144411, Punjab, IN

Abstract

 Objectives: The major focus of this research paper is to propose an Intelligent Authentication based Vehicle Initiated Broadcast-Dynamic Path (IAVIB-DP) data collection scheme with the aim of increasing the effectiveness of existing Vehicle Initiated Broadcast-Complete Path (VIB-CP) data collection scheme which is considered to be the best and most commonly used way that is opted for data collection in VANET and is evaluated in terms of packet delivery ratio, average latency and communication overhead. Methods/Analysis: Simulation is conducted by using OMNet++ to compare the performance of IAVIB-DP with one of the best data collection scheme working on single RSU, VIB-CP. Performance Index (PI) is measured by evaluating the Packet Delivery Ratio (PDR), average latency and Communication Overhead (CO) for proposed and existing scheme of collecting data. Best scheme will be decided on the basis of calculated PI. Other parameters for simulation such as the minimum speed, space dimensions and the maximum speed of moving vehicles remain fixed. Findings: VIB-CP and IAVIB-DP Data Collection Schemes (DCSs) are compared and analysis is done on the calculated value of Performance Index. PI decides whether a data collection scheme is effective or not. PI is calculated on different factors like PDR, Latency and CO. The simulation results show that PI of proposed IAVIB-DP data collection scheme is more as compared to PI of VIBCP, as it has high PDR, low latency and less CO. Application/Improvements: VIB-CP and IAVIB-DP are compared on same factors and are used for PI calculation. On the basis of the simulation results it is evaluated that IAVIB-DP performs better.

Keywords

Communication Overhead (CO), Data Collection Schemes (DCSs), Packet Delivery Ratio (PDR), Performance Index (PI), Road Side Unit (RSU)

Full Text

   

Sarneet Kaur ¹, Babita Pandey ¹

Affiliations
1 Department of Computer Applications, Lovely Professional University, Jalandhar – 144411, Punjab, IN

Abstract

Protein secondary structure prediction plays very important role in prediction of tertiary protein structure which in turn gives accurate information about its functions. In this work we have used Feed Forward and Perceptron Neural Network to predict protein secondary structure and also compare their performances. In this method an integer value is assigned to each amino acid and the data is normalized between 0 and 1. The data is trained with window size of 3 using MATLAB. This method gives high accuracy of prediction of protein secondary structure. In future we will try to incorporate physicochemical properties of the amino acids.

Keywords

Feed Forward, Perceptron, Protein Secondary Structure.

Full Text