Search

Role of gene expression in cancer and cellular process is a complex problem that continues to haunt and challenge researchers. Sheer number of genes and inter related biological processes make the process of identifying more complex. Gene classification and analysis is a very difficult task for data scientists, as it requires various data, information and facts for making different quality articulation. Latent Dirichlet Allocation (LDA) has shown its compatibility for investigation and quality articulation in solid and malignant growth tissues. LDA is used to connect or grouping of genomic data and results are subsequently used for investigation. This work proposes A Modified Latent Dirichlet Allocation (MLDA) for gene classification and achieves quality articulation. The proposed MLDA identify and group differentially expressed genes between healthy and cancer tissues of various types. Experimental results report better performance of MLDA as compared to current state of art methods over Breast and Lung cancer data sets.

Mobile ad-hoc network is a dynamic and self configuring network composed of mobile nodes that are cooperative and intelligent in nature, which forms a temporary network without any base station. Due to dynamic nature, route among source and destination is not fixed and it can change with time, this results in routing overhead, congestion and higher energy consumption in nodes. In this condition, clustering becomes very relevant in highly dense network in which the clusterhead initiates the routing instead of normal nodes. This paper proposes an efficient way for cluster formation and selection of stable cluster heads (E-CFSA) with high energy level. E-CFSA forms node cluster using k-means algorithm where distance with centroids acts as the key parameter. Afterwards, artificial neural network (ANN) is applied in each cluster to select efficient clusterhead. It also updates weight of input parameters such as mobility, packet drop, energy and number of neighbor nodes in order to minimize the errors at target neuron using back propagation algorithm. The overall process identifies and selects the nodes as clusterheads that possess longer stability and higher energy level. This procedure reduces the repetitive selection of cluster heads and re-affiliation of member nodes in a cluster. Thereafter, performance is evaluated in terms of overhead, packet delivery ratio, throughput and cluster head stability time with variation in speed of node or number of nodes in MANET. The experimental results obtained through proposed E-CFSA are compared to other existing mechanisms and a comparison is drawn. These comparative results very clearly revealed that proposed E-CFSA outperformed its peers.

Medical diagnosis and extracting patterns that can be converted into useful knowledge is a challenging work for researchers. Medical records are based on real-time data that have high dimensionality which makes task of extracting patterns even more complex. Prediction of various diseases often suffers due to large dimensionality that includes thyroid, diabetes, cancer, etc. Data mining is a process of finding out information and constructing a knowledge base from a huge amount of data, which remains useful. Data mining achieves these tasks involving classification, clustering, regression, and prediction. Reduction of dimensionality with the motive of knowledge discovery is an essential aspect of data mining as it helps in prediction/determination for various data analytics filed like medical diagnosis, business modeling, and government data analysis. This paper presents cooperative method for classification of thyroid using machine learning fusing C4.5 and random forest classification technique (CCTML). Experimental results are compared with other conventional techniques and the results obtained through CCTML reported better classification accuracy and outperform conventional methods.

Self-tuning and detector power is introduced to address improper coverage of NSA.Cooperation among agents makes correct identification of anomaly and lower FAR.EPAADPS improves anomaly detection and reduces FAR compared to state of art.Preventive mechanism identifies and contains novel and zero day attacks. Artificial Immune System (AIS) is inspired from Biological Immune System (BIS) and demonstrates a lot of interesting facets and intelligence that include self-learning, self adaption, self regulatory, distributed with self/non-self detection capabilities. Due to these astonishing qualities AIS are predominantly used in anomaly detection where anomalies are treated as non-self that needs to be detected. Therefore, AIS appears appropriate for development of a proactive system to identify and prevent novel and unseen anomalies. This paper presents "An Efficient Proactive Artificial Immune System based Anomaly Detection and Prevention System (EPAADPS)" which embodies immune attributes to distinguish self and non-self in quest to identify and prevent novel, unseen anomalies. Negative Selection Algorithm (NSA) is a key AIS concept and is used for anomaly detection in various publications. Despite its relative success, detector selection and thereafter anomaly detection demands a more effective algorithm. This paper put forwards concept of self-tuning of detectors and detector power in NSA with the intension to make a detector evolve and facilitate better and correct self and non-self coverage. Thereafter, agents accompanying detectors collaborate and communicate between themselves to proactively discover correct anomalies and then take appropriate preventive measures. The performance of EPAADPS is contrasted with closely related state of art RNS algorithm using real valued representation and Euclidean distance. Experimental results revels promising EPAADPS performance which very comfortably outperforms the RNS. Furthermore, these results also demonstrate that EPAADPS shows remarkable resilience and intelligence in detecting novel unseen anomalies and with preventive measures to overcome the threat perception.

Image compression goal is to reduce the size of an image but without losing the information, the loss of information some time is acceptable but not always, today’s world the compression of image, video done with various techniques for the variety of an application. Medical science is one of the main application area for the image compression in which we compress the human body part images without losing the information and use in future. In this paper we proposed a hybrid model for the Bio-medical image compression which is better in the terms of result by measuring performance evaluation parameters to improve the value of performance parameter.

Image enhancement is one of the key concerns pertaining to better quality image photography captured through modern digital cameras. Probability of digital images getting compromised through lightning and weather conditions remains high. Due to these environmental limitations, many a time loss of information from images is reported. Major role of image amplification is to bring out hidden details of an image from the sample. It provides multiple options for enhancing the visual quality of images. This paper addresses problem of early skin cancer detection using image enhancement techniques and presents a multi-scale retinex with color restoration (MSR-CR) technique for skin cancer detection. The actual skin portion suffering from cancer is identified by comparing enhanced image with available ground truth image. Experimental result shows significant improvement over previously available techniques.

Content-based image retrieval based on color, texture and shape are important concepts that facilitate quick user interaction. Due to these reasons, humongous amount of explores in this direction has been done, and subsequently, current focus has now shifted in improving the retrieval precision of images. This paper proposes a dominant color and content-based image retrieval system using a blend of color, shape, and texture features. K-dominant color is extracted from the pixels finding and can be gathered in the form of cluster or color clusters for forming a cluster bins. The alike colors are fetched on the basis of distance calculations between the color combinations. Then the combination of hue, saturation, and brightness is calculated where hue shows the exact color, and the color purity is shown by saturation, and the brightness of the percentage degree increases from black to white. Experimental results clearly indicate that the proposed method outperforms the existing state of the art like LBP, CM, and LBP and CM in combination.

Human skin detection strives to spot skin from the pictures. Automatic skin detection is considered as a significantly difficult and complex as skin image differ on the aspects of contents due to variation in size, style, orientation, alignment coupled with different contrast and background. This paper proposes a skin detection approach using localization, tracking, extraction, enhancement, and recognition. This approach remains sensitive to the color palette and uses edge detection technique. Also, color classification box incorporates a deep impact on the performance of the rule. The proposed approach detects single as well as multiple persons in a picture. Promising results are obtained on variety of pictures, except in few pictures wherever color distinction is hard to even when edge detection rule.

Hand gesture recognition system is employed to provide interface between computer and human using hand gesture. This paper presents a technique for human computer interface through common hand gesture that is efficient to commemorate 25 aspersion gestures from the American sign language hand alphabet. The prospect of this paper is to develop up an algorithm for hand gesture recognition with reasonable accuracy. This work uses a domain independent learning methodology to automatically stir low-level spatio-temporal descriptors for high-level cross recognition by Correlated variance programming. Feature extraction is the most important orientation for gesture recognition and is indeed important in terms of giving input to a classifier. In this work Canny edge detector algorithm is used to find edge of the segmented and morphological filtered image which yields boundary of hand gesture in the image then Correlated variance mean based programming applied for recognition of gesture. Experimental results very precisely indicate that the developed method outperforms the existing state of the art.

Immune System (AIS) over the years has caught attention of researchers of various domains for complex problem solving. AIS model the procedure and methodologies of Biological Immune System (BIS) which protects the body from diverse attacks and different challenges. Scientists over the years are amazed with the appealing features of BIS that can be exploited. The most significant of them is its ability to distinguish self and non-self. This theory forms the basis of Negative Selection Algorithm (NSA) in AIS. NSA is competent for anomaly detection problems. From this perspective this research paper presents a Novel Immunity inspired approach for Anomaly Detection (NIIAD) with the feature of fine tuning. The main intention of adopting finetuning is to covering more self region and identifying non self region proficiently. Experimental results reflects high detection ratio with less false alarm and low overhead.

Danger Theory is presented with particular predominance on analogies in the Artificial Immune Systems world. Artificial Immune System (AIS) is relatively naive paradigm for intelligent computations. The inspiration for AIS is derived from natural Immune System (IS). The idea is that the artificial cells release signals describing their status, e.g., safe signals and danger signals. The various artificial cells use the signals in order to adapt their behavior. This new theory suggests that the immune system reacts to threats based on the correlation of various (danger) signals and it provides a method of ‘grounding’ the immune response, i.e., linking it directly to the attacker. In this paper, we look at Danger Theory from the perspective of AIS practitioners and an overview of the Danger Theory is presented with particular emphasis on analogies in the Artificial Immune Systems world.

Classifying high-dimensional datasets is a very challenging task. In high dimensional spaces, the performance of supervised learning methods suffer due to the huge difference between number of columns and number of rows, which degrades both accuracy and efficiency performance of classification. In this paper we propose a two phase approach to the given datasets to address the present scenario. Dimension Reduction Method Supervised Locally Linear Embedding (SLLE) is applied in the first phase to reduce dimension of the datasets and in next phase classification through K-NN and SVM is done. Experiments are carried on different high dimensional datasets and then we compared of different dimension reduction and classification methods.

DNA Micro arrays allow the parallel monitoring of thousands of expression levels of genes simultaneously. The secret behind of this technology is the fact that DNA nucleotide bases will hybridize to certain other nucleotide bases. With the help of DNA Micro arrays we can predict Human cancer considering the genes expression levels across a collection of sample. In this paper we introduce ensemble approach to classification method which included Dimension Reduction. For Dimension Reduction we use Supervised Locally Linear Embedding and for Classification uses Support Vector Machine.