A5064004667- Gene expression and cancer classification
- Face and Expression Recognition
- Data Mining Algorithms and Applications
- Artificial Intelligence in Healthcare
- Brain Tumor Detection and Classification
- Software Engineering Research
- Software Engineering Techniques and Practices
- Vehicle License Plate Recognition
- Imbalanced Data Classification Techniques
- Rough Sets and Fuzzy Logic
- Machine Learning and ELM
- Advanced Steganography and Watermarking Techniques
- Software Testing and Debugging Techniques
- COVID-19 diagnosis using AI
- Software System Performance and Reliability
- Digital Media Forensic Detection
- Anomaly Detection Techniques and Applications
- Evolutionary Algorithms and Applications
- Radiomics and Machine Learning in Medical Imaging
- Biometric Identification and Security
- vaccines and immunoinformatics approaches
- Spam and Phishing Detection
- Network Security and Intrusion Detection
- Handwritten Text Recognition Techniques
- Text and Document Classification Technologies
Amrita Vishwa Vidyapeetham
2012-2023
Agency for Science, Technology and Research
2007
Bioinformatics Institute
2007
Microarray gene expression data plays a prominent role in feature selection that helps diagnosis and treatment of wide variety diseases. contains redundant genes high dimensionality smaller training testing samples. This paper proposes customized similarity measure using fuzzy rough quick reduct algorithm for attribute selection. Information Gain based entropy is used to reduce the first stage proposed method defines selecting minimum number informative removing employed at second stage. The...
This Letter proposes a customised approach for attribute selection applied to the fuzzy rough quick reduct algorithm. The unbalanced data is balanced using synthetic minority oversampling technique. huge dimensionality of cancer reduced correlation-based filter. gene subset used compute final minimal set triangular norm operator on with Lukasiewicz implicator approximation. selects least number informative feature genes from datasets. Classification accuracy leave-one-out cross validation...
The main objective of this paper is to remove the redundant genes present in samples and thereby increase classifier accuracy. This accomplished by devising a hybrid approach for feature selection that selects subset from raw dataset then classifies based on training imparted. In paper, rank information gain filter used dimensionality reduction. Fuzzy rough set genetic algorithm methods were combined form prominent removes ones. process classification performed using extreme learning...
In recent times a lot of work has been carried out in the field reversible data hiding (RDH) to prevent secret from theft, illegal copying and unlawful reproduction. RDH cover image will be recovered after extracting which was embedded that image. Finding best location hide is an important task so it conceal existence message. This paper provides technique based on Firefly algorithm (FA). The optimal found by firefly algorithm. histogram shifting used embed Histogram Techniques have...
In this study, we applied a novel method by using correlation coefficient filter for dimensionality reduction followed fuzzy rough quick reduct algorithm feature selection. The classification performance was evaluated the gene subsets obtained from based and our proposed method. Later compared results with other traditional classifier techniques. After suitable experimental analysis, it has been found that two-fold advantage namely selection of much lesser number genes to improved accuracy...
The innate immune system is fundamental to the recognition of pathogens, triggering immune-inflammatory response and host defense. Recent advance in this area has resulted enormous amount data, which are stored across different databases. Integrating relevant information from these data sources difficult because their heterogeneous nature dispersed physical location. We present here a single portal system, Cell Interaction Knowledgebase, with focus on immunity. In particular, knowledgebase...
Cancer diagnosis is one of the emerging applications in microarray gene expression data. Feature selection plays a crucial role because huge dimensionality and less training testing samples. Finding small subset significant genes from large set data challenging task. This paper presents usage genetic algorithm as tool to determine informative uses Extreme Learning Machines classifier accuracy. Experiments are carried out on two datasets results reveal that proposed approach produces better...
Among the diverse applications of computer and communication technologies, Intelligent Transport System aids in simplifying transport problems. Its aim is to gather data provide timely feedback traffic managers (traffic policemen) road users. The various problems involved processing real-time has been addressed several areas research that includes vehicle detection, tracking classification. This paper proposes a technique for isolation classification vehicles at an abstract level. aims...
Diagnosis of cancer is one the most emerging clinical applications in microarray gene expression data.However, classification on data still remains a difficult problem.The main reason for this significantly large number genes present relatively compared to available training samples.In paper, novel approach feature extraction combining statistical t-test and absolute scoring proposed achieving better rate.Suitable approaches using linear Support Vector Machines, Proximal Machines Newton also...
Magnetic Resonance Imaging (MRI) images of the brain are used for detection various diseases including tumor. In such cases, classification MRI captured with respect to ventricular and eye ball regions helps in automated location diseases. The methods employed paper can segregate given into region region. First, image is segmented using Particle Swarm Optimization (PSO) algorithm, which an optimized algorithm segmentation. proposed then applied on image. detects whether consist a or...
Abstract: Farming is the cultivation of plants and livestock. Plant monitoring one most important tasks in farming. The goal this paper to use IoT NodeMCU system platform for plant smart gardening. primary reduce direct interaction provide comfort farmer by improving system's overall performance. Humidity, sunlight, soil moisture are factors consider when productivity. growth health information must be provided user on a continuous basis recording these parameters. interfaces with all...