Ninni Singh

Missing Value Imputation with Unsupervised Kohonen Self Organizing Map

Many data mining and data analysis techniques function with large datasets. These large data sets have missing values which result in biased estimates, imprecise statistical results or unacceptable conclusions. Data mining and data analysis techniques cannot be directly applied to datasets with missing values. For this purpose, different imputation techniques are proposed by different authors for both categorical and continuous variables. The existing imputation techniques have many limitations such as (a) methods like conditional mean imputation results in biased parameter estimation. (b) Too much variation is discovered in the inference of any single value or distance between particular samples in the case of random draw imputation. (c) In case of multiple imputations it is not easy to determine the posterior distribution of samples to draw from. In this paper, we present an unsupervised learning technique based on a Kohonen self-organizing map used for both categorical and numerical data values. In this paper, our aim is to achieve the highest accuracy. To achieve this, we trained our model by using the splitting approach to make the learning model and use this model to predict the accuracy. The proposed algorithm can map the missing values closed to original by adjusting the weights by improving accuracy when compared to classification without missing values and with missing values.

Image data-deduplication using the block truncation coding technique

In this paper, we have CBIR, which is known as Content based Image Retrieval is the application of computer vision technique to the image retrieval problem from the large databases. In this paper de-duplication of electricity bills is done through CBIR. The technique for comparing the images is Block truncation coding; it is a lossy compression technique for grayscale image. So Photograph of electricity was divided into different block size using a compression technique known as Block Truncation coding. On the basis of block size, we will put images of the same block size in the same cluster. As there will be large block, so they will be divided into chunks of data. Through this experiment we will try to remove duplicates in the same cluster. In this paper the block of data is also divided into different chunks, so that image of single instance can be stored in these chunks. So duplication will be improved with the help of compression techniques discussed in detail.

A Secure Authentication Scheme in Multi-operator Domain (SAMD) for Wireless Mesh Network

Wireless mesh network (WMN) is considered to be an evolving technique because of self-configuration and adaptive features, it supports large-scale network especially in an organization and academics. As with any network, communication among nodes plays an important role, when two nodes in a network communicate with each other via the Internet, secure authentication is an imperative challenge. In literature, there are many approaches that have been suggested to deliver a secure authentication between nodes in wireless mesh network (WMN), however, all these outlines contain some disadvantages, i.e. management cost of the public key and system complexity. Our suggested proposed approach is dealing with one of the wireless mesh network challenges, i.e. Mutual authentication. Here in this we have considered the three authentication techniques, i.e. Inter-, Intra- and Inter-Operator domain authentication. SAMD (Secure Authentication Scheme in Multi-Operator Domain) hides the location, communication path and network access and apart from that it will resist to the adversary, key forgery and modification attack. As a result, we have assessed the performance of the proposed design in terms of Authentication cost, Encryption cost, Key validation, Key Generation, Throughput and System Delay, which indicates our scheme more efficient than other schemes. Results show that it will efficiently work in real-time traffic.

Prediction of Reactor Performance in CATSOL-Based Sulphur Recovery Unit by ANN

For the prediction Artificial Neural Network has been used, which is inspired from human biological neural network. Artificial Neural Network tool has very vast application in almost every field. ANN is a very powerful tool in the field of AI and spreading over in every field of Computer Science, IT, Refinery, Medical, etc. We have used MATLAB software for the use of built-in neural network toolbox. MATLAB platform is very user friendly and includes all the important transfer functions, training functions which helps in comparison of results with each other and allow us to predict the best result using ANN. Further, this describes application of ANN in CATSOL process. After complete study about ANN and CATSOL process, our experimental results of using different transfer functions and training functions are shown which results in the overall best network for the CATSOL process.

Development of Knowledge Capsules for Custom-Tailored Dissemination of Knowledge of Seismic Data Interpretation

Knowledge management system is a repository of factual information. Seismic data interpretation is a field of exploration geophysics, which deals with interpretation of seismic images, to infer subsurface geology and provide information regarding hydrocarbon accumulation. This knowledge of interpretation is rare, expensive and largely individualistic. Lack of formal interpretation rules, causes seismic experts to use their own expertise gained over years of experience, leading to uncertainty. In current work a knowledge management framework is proposed, which initiates with the knowledge engineer gathering tacit knowledge from seismic experts, followed by a knowledge manager, synchronizing, sequencing, formalizing and organizing it in explicit form to develop a knowledge capsule, to facilitate its sharing through tutoring. Knowledge capsules have been refined to effectively suit different levels and knowledge grasping preferences of novice seismologists.

Formal Verification of Secure Authentication in Wireless Mesh Network (SAWMN)

Wireless mesh network (WMN) is considered to be an evolving technique because of self-configuration and adaptive features, and it supports large-scale network especially in an organization and academics. As with any network, communication among nodes plays an important role, when two nodes in a network communicate with each other via the internet, secure authentication is an imperative challenge. In literature, there are many approaches that have been suggested to deliver a secure authentication between nodes in WMN; however, all these outlines contain some disadvantages, i.e., management cost of the public key and system complexity. In this paper, a Secure Authentication in Wireless Mesh Network (SAWMN) approach is proposed which overcomes these drawbacks and provides an efficient authentication to the mesh clients. Further, SAWMN results have been shown simulated on AVISPA SPAN to ascertain the authenticity of the proposed approach.

Digital Watermarking using Discrete Wavelet Techniques with the help of Multilevel Decomposition Technique

Digital watermarking plays an important role for securing data, providing authentication, confidentiality and even protects digital images from illegal distortion. In this paper, we are particular using discrete wavelet transform which is more prevalent than others. The watermark logo is inserted after compressing the image with the help of JPEG compression technique. First original image is taken; it is changed into multilevel decomposed with the help of HAAR wavelet, as it is a type of wavelet used for diet (discrete wavelet technique) technique. The algorithm has been tested under the presence of attacks like Gaussian noise, JPEG compression and we see how their BER and PSNR value varies. Further, compression before watermarking is done to ensure inaccessibility of information to unauthorized personnel.