Bhupesh Gour

A Semi- Supervised Technique for Weather Condition Prediction using DBSCAN and KNN

Weather condition prediction has always been a keen area of interest among researchers and climate change prediction experts. Due to gradual changes in the atmospheric and climatic conditions the appropriate prediction task has become a formidable challenge. In this paper we propose a semisupervised weather prediction technique to validate the predictions done for certain atmospheric parameters taken for four years on a day wise basis in a certain city. The experimental outcomes of this work show that this semi supervised technique provides appropriate results and can be used for weather condition prediction & analysis.

A Weighted Markov Model for Web Pre-fetching to Improve User Interface over Internet

------------------------------------------------------------------------ABSTRACT-------------------------------------------------------------Rapid growth of web application has increased the researcher’s interests in this era. All over the world has surrounded by the computer network. There is a very useful application call web application used for the communication and data transfer. An application that is accessed via a web browser over a network is called the web application. Web caching is a well-known strategy for improving the performance of Web based system by keeping Web objects that are likely to be used in the near future in location closer to user. The Web caching mechanisms are implemented at three levels: client level, proxy level and original server level. Significantly, proxy servers play the key roles between users and web sites in lessening of the response time of user requests and saving of network bandwidth. Therefore, for achieving better response time, an efficient caching approach should be built in a proxy server. This paper use FP growth, weighted rule mining concept and Markov model for fast and frequent web pre fetching in order to has improved the hit ratio of the web page and expedites users visiting speed.

An Extensive Estimation and Analysis of Image Denoising

Image denoising is the technique of removal of the noise from the image contaminated by additive Gaussian noise without loss of features of image. It is a fundamental process in pattern recognition and image processing. When an Image is captured many factors such as lighting spectra, source, intensity and camera Characteristics affect the image. The main factor that reduces the quality of the image is Noise. It hides the important information of images and changes value of image pixels at key locations causing blurring and various other deformities. Noises must be removed from the images without loss of any information with it. Noise removal is the preprocessing stage of image processing. There are many types of noises which may corrupt the images. These noises are appear on images in many ways: at the time of acquisition due to noisy sensors, due to defective scanner or due to faulty digital camera device, as a result of transmission channel errors, due to corrupted storage media. There are numerous researches have been done on wavelet based denoising for estimation of parameters such as variance of the multi scale Linear minimum mean square error. In this review paper we have presented an extensive analysis and literature review on image denoising.

Analysis of visibility and temperature patterns of Indian cities and it's clustering to identify the effect of presence of aerosol particles in the atmosphere

Variations in ambient air quality data are caused by changes in the pollutant emission rate, and meteorological and topographical conditions of the place. Mass concentration of aerosol is a measure of air quality and aerosol source strength at a particular location. It has been shown that clear sky visibility over land has decreased globally over the past 30 years, indicative of an increase in aerosols, or airborne particulates, over the worlds continents during that time. The change in climatic conditions is of great concern in environment, industry and agriculture. The disturbance of temperature and other climate factors due to presence of aerosol particles in air, results in global climate changes. The aim of this research is to develop artificial neural network based clustering method for ambient atmospheric condition prediction in Indian city. Self-Organizing Map (SOM) Neural Network to divide data into four clusters which represents association in between atmospheric conditions belonging to cities of one cluster due to the amount of aerosol particles present in the atmosphere of those cities. The experimental results determined climate changes due to concentration of aerosol particles in the atmosphere of different cities in India and the correlation in between change in visibility and change in the temperature during the months of March to June.

Implementation of Different Ant based Techniques for Network Load Analysis

Network Load balancing is a technique of balancing at each node the number of packets received and the number of packets forward to the other node so that the chance of network congestion problem has been reduced and bandwidth is utilized. Although there are many techniques implemented for the balancing of nodes based on maintaining a routing table at each node and is updated as the packet get forward from that node. Ant Colony Optimization is one of the techniques used in the network for the balancing of number of packets at each node. Here in this paper is proposed a comparative study of different ant colony optimization techniques implemented for the analysis of the network load balancing. Here the ant based techniques are implemented are simulated for different conditions and on the basis of which proposed the best ant based techniques for the network load balancing.

Gain Comparison between NIFTY and Selected Stocks identified by SOM using Technical Indicators

The main aim of every investor is to identify a stock that has potential to go up so that the investor can maximize possible returns on investment. After identification of stock the second important point of decision making is the time to make entry in that particular stock so that investor can get maximum returns on investment in short period of time. There are many conventional techniques being used and these include technical and fundamental analysis. The main issue with any approach is the proper weighting of criteria to obtain a list of stocks that are suitable for investments. This paper proposes a method for stock picking and finding entry point of investment in stocks using a hybrid method consisting of self-organizing maps and selected technical indicators. The stocks selected using our method has given 37.14% better returns in a period of one and a half month in comparison to NIFTY.