Segmentation of Pectoral Muscle in Mammograms Using Granular Computing

In this paper, pectoral muscle segmentation was performed to study the presence of malignancy in the pectoral muscle region in mammograms. A combined approach involving granular computing and layering was employed to locate the pectoral muscle in mammograms. In most cases, the pectoral muscle is found to be triangular in shape and hence, the ant colony optimization algorithm is employed to accurately estimate the pectoral muscle boundary. The proposed method works with the left mediolateral oblique (MLO) view of mammograms to avoid artifacts. For the right MLO view, the method automatically mirrors the image to the left MLO view. The performance of this method was evaluated using the standard mini MIAS dataset (mammographic image analysis society). The algorithm was tested on 322 images and the overall accuracy of the system was about 97.47 %. The method is robust with respect to the view, shape, size and reduces the processing time. The approach correctly identifies images when the pectoral muscle is completely absent.

Mobility and Trajectory-Based Technique for Monitoring Asymptomatic Patients

Asymptomatic patients (AP) travel through neighborhoods in communities. The mobility dynamics of the AP makes it hard to tag them with specific interests. The lack of efficient monitoring systems can enable the AP to infect several vulnerable people in the communities. This article studied the monitoring of AP through their mobility and trajectory towards reducing the stress of socio-economic complications in the case of pandemics. Mobility and Trajectory based Technique for Monitoring Asymptomatic Patients (MTT-MAP) was established. The time-ordered spatial and temporal trajectory records of the AP were captured through their activities. A grid-based index data structure was designed based on network topology, graph theory and trajectory analysis to cater for the continuous monitoring of the AP over time. Also, concurrent object localisation and recognition, branch and bound, and multi-object instance strategies were adopted. The MTT-MAP has shown efficient when experimented with GeoLife dataset and can be integrated with state-of-the-art patients monitoring systems.

Automatic Text Summarization by Providing Coverage, Non-Redundancy, and Novelty Using Sentence Graph

The day-to-day growth of online information necessitates intensive research in automatic text summarization (ATS). The ATS software produces summary text by extracting important information from the original text. With the help of summaries, users can easily read and understand the documents of interest. Most of the approaches for ATS used only local properties of text. Moreover, the numerous properties make the sentence selection difficult and complicated. So this article uses a graph based summarization to utilize structural and global properties of text. It introduces maximal clique based sentence selection (MCBSS) algorithm to select important and non-redundant sentences that cover all concepts of the input text for summary. The MCBSS algorithm finds novel information using maximal cliques (MCs). The experimental results of recall oriented understudy for gisting evaluation (ROUGE) on Timeline dataset show that the proposed work outperforms the existing graph algorithms Bushy Path (BP), Aggregate Similarity (AS), and TextRank (TR).

Gradient Boosting Machine and Deep Learning Approach in Big Data Analysis

Designing a system for analytics of high-frequency data (Big data) is a very challenging and crucial task in data science. Big data analytics involves the development of an efficient machine learning algorithm and big data processing techniques or frameworks. Today, the development of the data processing system is in high demand for processing high-frequency data in a very efficient manner. This paper proposes the processing and analytics of stochastic high-frequency stock market data using a modified version of suitable Gradient Boosting Machine (GBM). The experimental results obtained are compared with deep learning and Auto-Regressive Integrated Moving Average (ARIMA) methods. The results obtained using modified GBM achieves the highest accuracy (R2 = 0.98) and minimum error (RMSE = 0.85) as compared to the other two approaches.

Deep Stacked Autoencoder-Based Automatic Emotion Recognition Using an Efficient Hybrid Local Texture Descriptor

Extracting an effective facial feature representation is the critical task for an automatic expression recognition system. Local Binary Pattern (LBP) is known to be a popular texture feature for facial expression recognition. However, only a few approaches utilize the relationship between local neighborhood pixels itself. This paper presents a Hybrid Local Texture Descriptor (HLTD) which is derived from the logical fusion of Local Neighborhood XNOR Patterns (LNXP) and LBP to investigate the potential of positional pixel relationship in automatic emotion recognition. The LNXP encodes texture information based on two nearest vertical and/or horizontal neighboring pixel of the current pixel whereas LBP encodes the center pixel relationship of the neighboring pixel. After logical feature fusion, the Deep Stacked Autoencoder (DSA) is established on the CK+, MMI and KDEF-dyn dataset and the results show that the proposed HLTD based approach outperforms many of the state of art methods with an average recognition rate of 97.5% for CK+, 94.1% for MMI and 88.5% for KDEF.

Phish-Shelter

Phishing attack is a deceitful attempt to steal the confidential data such as credit card information, and account passwords. In this paper, Phish-Shelter, a novel anti-phishing browser is developed, which analyzes the URL and the content of phishing page. Phish-Shelter is based on combined supervised machine learning model.Phish-Shelter browser uses two novel feature set, which are used to determine the web page identity. The proposed feature sets include eight features to evaluate the obfuscation-based rule, and eight features to identify search engine. Further, we have taken eleven features which are used to discover contents, and blacklist based rule. Phish-Shelter exploited matching identity features, which determines the degree of similarity of a URL with the blacklisted URLs. Proposed features are independent from third-party services such as web browser history or search engines result. The experimental results indicate that, there is a significant improvement in detection accuracy using proposed features over traditional features.

A Robust Authentication System With Application Anonymity in Multiple Identity Smart Cards

User Authentication plays a crucial role in smart card based systems. Multi-application smart cards are easy to use as a single smart card supports more than one application. These cards are broadly divided into single identity cards and Multi-identity cards. In this paper we have tried to provide a secure Multi-identity Multi-application Smart Card Authentication Scheme. Security is provided to user’s data by using dynamic tokens as verifiers and nested cryptography. A new token is generated after every successful authentication for next iteration. Anonymity is also provided to data servers which provides security against availability attacks. An alternate approach to store data on servers is explored which further enhances the security of the underlying system.

Crowd Abnormality Detection Using Optical Flow and GLCM-Based Texture Features

Detection of abnormal crowd behavior is one of the important tasks in real-time video surveillance systems for public safety in public places such as subway, shopping malls, sport complexes and various other public gatherings. Due to high density crowded scenes, the detection of crowd behavior becomes a tedious task. Hence, crowd behavior analysis becomes a hot topic of research and requires an approach with higher rate of detection. In this work, the focus is on the crowd management and present an end-to-end model for crowd behavior analysis. A feature extraction-based model using contrast, entropy, homogeneity, and uniformity features to determine the threshold on normal and abnormal activity has been proposed in this paper. The crowd behavior analysis is measured in terms of receiver operating characteristic curve (ROC) & area under curve (AUC) for UMN dataset for the proposed model and compared with other crowd analysis methods in literature to prove its worthiness. YouTube video sequences also used for anomaly detection.

Integration of Multi-Omics Data to Identify Cancer Biomarkers

A novel method for integrating multi-omics data, including gene expression, copy number variation, DNA methylation, and miRNA data, is proposed to identify biomarkers of cancer prognosis. First, survival analysis was performed for these four types of omics data to obtain survival-related genes. Next, survival-related genes detected in at least two types of omics data were selected as candidate genes. The four types of omics data only composed of candidate genes were subjected to dimension reduction using an autoencoder to obtain a one-dimensional data representation. The mRMR algorithm was used to screen for key genes. This method was applied to lung squamous cell carcinoma and 20 cancer-related genes were identified. Gene function analysis revealed that the genes were related to cancer. Using survival analysis, the genes were verified to distinguish between high- and low-risk groups. These results indicate that the genes can be used as biomarkers for cancer.

An Ensemble of Random Forest Gradient Boosting Machine and Deep Learning Methods for Stock Price Prediction

Stochastic time series analysis of high-frequency stock market data is a very challenging task for the analysts due to the lack availability of efficient tool and techniques for big data analytics. This has opened the door of opportunities for the developer and researcher to develop intelligent and machine learning based tools and techniques for data analytics. This paper proposed an ensemble for stock market data prediction using three most prominent machine learning based techniques. The stock market dataset with raw data size of 39364 KB with all attributes and processed data size of 11826 KB having 872435 instances. The proposed work implements an ensemble model comprises of Deep Learning, Gradient Boosting Machine (GBM) and distributed Random Forest techniques of data analytics. The performance results of the ensemble model are compared with each of the individual methods i.e. deep learning, Gradient Boosting Machine (GBM) and Random Forest. The ensemble model performs better and achieves the highest accuracy of 0.99 and lowest error (RMSE) of 0.1.