Intelligent Decision Technologies

Cloud computing is growing massively in the world of information technology. As a result of which the number of servers and virtual machines is increasing day by day. It leads to huge energy consumption and carbon emission. In this regard, task scheduling has drawn the attention of cloud service providers to reduce energy consumption and increase cloud utilization. This is a crucial issue in a heterogeneous cloud environment. In this paper, energy consumption issues of cloud datacenters are addressed and an Energy-Aware Cloud Task Scheduling (EACTS) algorithm is proposed. It is taking the concept from traditional min-min, max-min, and sufferage heuristics and merge with the energy model. These heuristics are implemented on the heterogeneous cloud environment. EACTS energy model focuses on the estimation of energy consumption in cloud datacenters. The proposed algorithms estimated the makespan, cloud utilization, and energy consumption for a benchmark dataset. The experimental results obtained from EACTS algorithm provides a valid arbitration between energy efficiency and makespan. It has shown the comparison between above mentioned algorithms for various scheduling parameters.

Chronic obstructive pulmonary disease (COPD) has been impacting a large population. It has a higher fatality rate than that of lung cancer. Diagnosis of this disease is quite challenging. Medical images analysis has been able to solve this challenge by early and accurate diagnosis of pulmonary disease. This analysis technique helps in pre-diagnosis and providing timely medical treatment thus reducing the mortality rate. The goal of this study is to establish an accurate process for classifying CT scan images into healthy lungs, COPD and Fibrosis impacted lung images. This classifying process has three steps. In the first step, lung scan is used for feature extraction. Then second and third step of feature selection and lung disease identification are carried using Machine Learning (ML) classifier.

Haralick texture features with Gray Level Co-occurrence Matrix (GLCM), Zernike’s moments, Gabor features and spatial domain features are used for feature extraction from the segmented lung CT images. For feature selection, our proposed evolutionary algorithm is the Improvised Grasshopper Algorithm (IGOA). After feature extraction from CT scan medical images, IGOA selects an optimal set of features that increases the classification accuracy and decreases the cost of computation. Lastly, three ML classifiers viz. Decision Tree Classifier, k-Nearest Neighbor (KNN), Random Forest Classifier are applied to every feature set chosen by IGOA.

The research results show that IGOA filtered out the maximum number of unimportant features of about 71.01%. IGOA eliminates 28.99% of the total extracted features. IGOA gave a better accuracy of 99.8%. Research results imply that the introduced feature selection method is appropriate for disease classification from CT scan images. IGOA method can be used for real-time applications as it has a less computational cost and has better accuracy.

In the era of electronic governance, electronic filing of import bills of entry documents and export bills documents is in place. All countries are thinking to eliminate the non-tariff barrier globally and the concept of liberalization is being introduced. Import duty is imposed when any imports take place. The drawback is considered as the incentive given to the exporters by the government of a country to encourage exports. The drawback is calculated based on the declared value of exported items, item-wise/quantity-wise. Many unscrupulous exporters overvalue the exported goods to get the higher incentive. So, determining the actual valuation of exported goods is a very challenging task for customs officials to prevent drawback frauds. Customs officials are in the process of valuation of exported goods with their discretion/skills from time to time for detecting drawback frauds and prevent government losses by scrutiny exports documents, invoices, and examining the goods. In this paper, a two-tier duty drawback fraud detection system using intuitionistic fuzzy theory and game theory. In tier-1, an intuitionistic fuzzy approach has been applied to determine the sensitivity of goods whereas, in tier-2, a game theory has been used to maximize the payoff of respective country governance.

During the last decade, a variety of ensembles methods has been developed. All known and widely used methods of this category produce and combine different learners utilizing the same algorithm as the basic classifiers. In the present study, we use two well-known approaches, namely, Rotation Forest and Random Subspace, in order to increase the effectiveness of a single learning algorithm. We have conducted experiments with other well-known ensemble methods, with 25 sub-classifiers, in order to test the proposed model. The experimental study that we have conducted is based on 35 various datasets. According to the Friedman test, the Rotation Forest of Random Subspace C4.5 (RFRS C4.5) and the PART (RFRS PART) algorithms exhibit the best scores in our resulting ranking. Our results have shown that the proposed method exhibits competitive performance and better accuracy in most of the cases.

During the 2 nd phase of COVID-19 pandemic, pharmaceutical plant industry is facing lot of production pressure and machine availability plays vital role in maximizing the manufacturing pharmacy product output. In this paper, Artificial Neural Networks (ANNs) based information processing algorithm has been used to provide a solution to this problem and it has been found suitable to predict machines availability as a prediction function. The considered pharmaceutical plants are dealing with production of medicines related common symptoms in case of COVID-19 (fever, coughing, and breathing problems). The pharmaceutical plant data corresponding to different values of repair and failure rates of different subsystems is collected from plant and analyzed with the help of validated neural network value of availability. This configuration of ANNs approach developed in this research allowed simplifying computational complexities of conventional approaches to solve a large plant machines availability problem. The ANNs methodology in the paper permitted making no assumption, no explicit coding of the problem, no complete knowledge of system configuration, only raw input and clean data found to be sufficient to determine the value of machine availability function for different value of failure and repair rates considered in the paper. The results obtained in the paper are useful for the plant leadership, as the value of failure and repair rates of various subsystems can be fine-tuned at a require clear-cut level to achieve higher availability, and avoid considerably loss of production, loss of man power, and by-pass complete breakdown of concerned system.

A new paradigm has been introduced by researchers in the field of data accessing at several mobile positions through implementation of WAN instead of cellular telephony. Based on range/coverage and network topology WAN taxonomy has been categorized as fixed network, mobile access network and ad – hoc network/wireless mesh networks respectively. Several interesting challenges has been posed by researchers in wireless networks such as routing, bandwidth, congestion, security, quality of service (QoS) and data rate enhancement. Of late, researchers around the world have started working in the domain of SHiP (Sparse, Hybrid, Intermittent and Partitioned) Network, which are plagued by several characteristics features such as: sporadic connections, intermittent routing, store – carry – forward type of message forwarding mechanism, dynamic network topology, transitive closure of message transmission, fluctuating network link, bottleneck probability, trust, and security in social networks. In this paper an outline of Oppnet (Opportunistic Network), part of SHiP Network has been presented with as in-depth coverage of routing problems and research challenges in it. Additionally, taxonomy of Oppnet routing protocols is illustrated with the features on basis of which categorization has been done. An exhaustive investigation through implementation of well-known routing schemes like: Epidemic, Spray and Wait, PROPHET and Direct delivery against mobility models such as: Random Walk, Random Waypoint, Shortest path-based movement and Cluster Movement has been accomplished in open-source software: ONE simulator with real world mobility dataset and traces of city of New Delhi, capital of India. Results achieved have been good and exceed our expectations.

Software products are essential parts of many organizations on-going business up to a large extent. The main factors contributing to the successful delivery of a software product are its timely completion within the allocated budget and its quality compliance. Customer goodwill and profitability are very important for a software organization’s continued business. A large proportion of software products are delivered late or go over-budget causing significant inconvenience to the customers. This work proposes an accurate development effort estimation approach for software products. The Class Point (CP) approach with regression analysis method has been used for estimation of the development effort. This work uses a two step estimation approach. In the first step, an enhanced CP approach is used to evaluate the development effort of the system. In the second step, regression analysis models are utilized to refine the estimated effort accuracy. The results derived by applying the proposed two step approach confirmed the validity and the accuracy of this approach. It was observed that the SVR with RBF kernel is providing the best accuracy compared to other approaches.

Face recognition systems have imprinted its presence in many applications from offices to security to daily use as in personal digital gadgets. With many Face recognition systems in use, still there is scope for its performance improvement. The performance of such systems suffers due to presence of covariates like non-uniform illumination, pose deviations, occlusions, low resolution etc. Extracting the region of interest from input face images is very crucial steps in face recognition system. We proposed methodology to create a bounding box for every sample face image. This bounding box dynamically tries to fit the face image to cover the relevant face features useful for recognition purpose. We used a light CNN structure to perform the experiment on MUCT face dataset using three different methodologies for extracting region of interests from sample face images using bounding box. These are 1) min-max based ROI 2) OpenCV face alignment and 3) eye-reference based ROI. It is observed that the proposed eye-reference based face alignment method works better than conventional methods of min-max based ROI and OpenCV face alignment with considerable amount of improvement in the recognition accuracy. Our future work includes use of other structured datasets with various covariates for face recognition.

Kinship Verification from facial images is known to have attracted major attention since time immemorial. Identifying the underlying patterns that exist between images and analysing the relationship hidden between them have enabled the multitudes of applications to utilize kinship relationships. This work serves as a study on the amount of influence that hereditary features can exert on the families tied together by lineage in identifying the relationship prevailing between them and whether it holds true or not. The approach employed involves detecting the relationship existing between the provided facial images using Siamese Network, which comprises two identical convolutional neural networks that share common weight values. A difference vector is computed from this Siamese CNN, which is then fed into a network of fully connected linear layers. This extended layer will determine whether the two individuals in the input images are related to each other or not.

In support of its faster learning capacity and better generalization, Extreme Learning Machine (ELM) has gained the attention of researchers as a means to solve various real-world prediction problems. However, the performance of ELM is heavily dependent on the activation functions used in it. In this study, design of ELM is addressed as a Multi Criteria Decision Making (MCDM) problem. The selection of the activation function for the ELM based predictor model is done through a novel MCDM ensemble approach. On the basis of 9 prediction metrics, MCDM techniques such as TOPSIS, PROMETHEE-II, and VIKOR were used to assess and rank 15 activation functions on ELM performance. In light of the fact that the ranks determined by each MCDM technique do not coincide, a novel ensemble approach was proposed to calculate the final rank score by considering the occurrences of each model in the primary ranking and its respective rank score. In the end, the most highly ranked activation function is taken into account in the ELM-based predictor model. The proposed model is assessed over three benchmark stock indices such as BSE SENSEX, NIFTY 50 and BSE S&P 500. The empirical analysis clearly shows that the ELM based predictor model designed using ELU activation function performs competitively compared to other reported models.

Plant diseases detection based on machine learning and computer vision can produce a significant effect on the quality and production of crops. Any changes that occur in crop quality or crop productivity may greatly reduce the national economy. Thus, the detection of plant diseases should be done at the early stage before intensively affecting crop production. A new technique named smart farming is introduced to benefit in “high-ended application of modern farming” by obtaining multiple data through live streams, social media, sensors, robots, etc. The attained data from diverse sources are required to processunder amultilevel database, which becomes more challenging while detecting plant diseases in smart farming techniques. The demands for using the machine learning approaches with unsupervised or supervised methods are increased on utilizing it in real-world applications. The main intention of this paper is to focus on the development of a novel crop disease detection model using the modified deep learning architecture. The images from different datasets with several crop diseases are collected from the public benchmark sources, and it is initially subjected to pre-processing using filtering and contrast enhancement techniques. Once the image is enhanced, the novel Optimized K-means Clustering (O-KMC) is adopted for performing the abnormality segmentation. Then, the feature extraction of the abnormality segmented images is done by the edge features and texture features. These features are utilized for disease recognition, in which the Heuristic-based Convolutional Neural Network with Recurrent Neural Network (H-C-RNN) is developed. In both segmentation and classification, the parameter improvement is performed by the Adaptive Inertia Weighted-Dragonfly Algorithm (AIW-DA). The performance of the proposed model under the different datasets is evaluated with various conventional methods that ensure the accurate identification of crop diseases in the proposed model.

The first version of the RLGame ecosystem, which features a collection of conventional Artificial Intelligence and Reinforcement Learning techniques for learning to play (and, actually, playing) a board game, was adopted for the development of the second version that supports Multiagent Reinforcement Learning and a sequence of games. We present an experimental comparison of a variety of algorithms which are available within the ecosystem and comment on the potential to investigate single-player vs multi-player scenarios, alongside some experimental results which suggest that loosely co-ordinated game play can be superior to fully co-ordinated game play.

Biomarker plays an important role in early disease diagnosis including cancer. The World Health Organization defines a biomarker as any structure or process in the body that is measurable and affects the prognosis or outcome of the disease. Today, biomarkers can be identified using bioinformatics tools. The detection of biomarkers in the field of bioinformatics is considered more as a problem of feature selection. Many feature selection algorithms have been used for biomarker discovery however these algorithms do not have enough accuracy or have computational complexity. For this reason, the researchers discard the high accuracy algorithms because they are time consuming. We redesigned an efficient algorithm based on parallel algorithms. We used the Cancer Genome Atlas (TCGA) including breast cancer patients. The proposed algorithm has the same accuracy and increases the speed of algorithm.

Smart grid systems are being actively developed and implemented all over the world. However, along with developed systems for monitoring and data analysis, decision support functions are not fully implemented. Wherein decision support is necessary due to the complexity of possible emergencies. In this work, we offer the concept of an intelligent decision support system (IDSS) for the SMART grid, which is based on the hybrid Case-Based Reasoning (CBR) method. This method combines models of knowledge-based systems and models of neural networks and machine learning, which simplifies realization on complex changing objects of the SMART grid. In the first part of the research, we describe the concept of the proposed hybrid-CBR method, the principle of formalizing the situation at the objects of the SMART grid systems and present the involved neural network architecture Comparator-Adder. The second parts of the research reveal the concept of applied IDSS and also show the results of an experiment of retrieving precedent from a knowledge base with using a neural network. Experimental results show that our architecture successfully copes with the task of selecting the most similar situation. We believe that the MAPE error in this incident does not play a key role; the efficiency of the neural network is confirmed primarily by the coherence with the results of the expert choice and the absence of collisions.

A cognitive-analysis of facial features can make facial expression recognition system more robust and efficient for Human-Machine Interaction (HMI) applications. Through this work, we propose a new methodology to improve accuracy of facial expression recognition system even with the constraints like partial hidden faces or occlusions for real time applications. As a first step, seven independent facial segments: Full-Face, half-face (left/right), upper half face, lower half face, eyes, mouth and nose are considered to recognize facial expression. Unlike the work reported in literature, where arbitrarily generated patch type occlusions on facial regions are used, in this work a detailed analysis of each facial feature is explored. Using the results thus obtained, these seven sub models are combined using a Stacked Generalized ensemble method with deep neural network as meta-learner to improve accuracy of facial expression recognition system even in occluded state. The accuracy of the proposed model improved up to 30% compared to individual model accuracies for cross-corpus seven model datasets. The proposed system uses CNN with RPA compliance and is also configured on Raspberry Pi, which can be used for HRI and Industry 4.0 applications which involve face occlusion and partially hidden face challenges.