Machine Learning (ML)

One of the greatest challenges of any system is the efficient allocation of resources. During any pandemic, even well organized medical systems face many issues to facilitate patients in an appropriate way. This paper will present the survey for the usage of intelligent technologies in the allocation and management of resources in hospitals. It will conclude with the scope of technologies that can be used during COVID-19 for better hospital resource management.

Artificial intelligence (AI) is one of the leading trends in modern day education system. The prospect of AI in modern day education system is very much important and significant. This paper deals with the different prospects of artificial intelligence (AI) in terms of education process. In this present scenario, AI is being used for the development and progress of teaching learning process which also makes the education easy accessible and effective. Learners are becoming more interested in the use of AI oriented education process. The researcher has also tried to study the future prospects of AI in education system. How the global education system is gradually becoming dependable on the use of AI is also the area of study. Till now AI aren’t capable of replacing teachers but in the upcoming future education sector will be of no use without AI. The present study addresses some genuine and potential aspects of AI in the field teaching learning. The present study is designed in the form of scheduled interview where samples were randomly selected from various schools of West Bengal. The reports of the interviews are studied and understood in the form of case study.

Parse details from a resume using natural language clarifying, find the keywords, assemble them onto sectors based on their keywords and lastly show the most relevant resume to the manageress based on keyword matching. Initially, the user transmits a resume to the web platform. The parser parses all the crucial information from the resume and auto fills a form for the user to analyze. Once the user confirms, the resume is displayed to the employers. Also, the user gets their resume in both JSON format and GUI view. Parsed data include name, email address, social profiles, years of work experience, work experiences, years of edification, education experiences, publications, certifications, volunteer experiences, keywords and finally the cluster of the resume , Computer science, human resource are the best examples.

Comparison study of algorithms is very much required before implementing them for the needs of any organization. The comparisons of algorithms are depending on the various parameters such as data frequency, types of data and relationship among the attributes in a given data set. There are number of learning and classifications algorithms are used to analyse, learn patterns and categorize data are available. But the problem is the one to find the best algorithm according to the problem and desired output. The desired result has always been higher accuracy in predicting future values or events from the given dataset. Algorithms taken for the comparisons study are Neural net, SVM, Naïve Bayes, BFT and Decision stump. These top algorithms are most influential data mining algorithms in the research community. These algorithms have been considered and mostly used in the field of knowledge discovery and data mining.

The universe is a gigantic ever-expanding mess. To classify it, Is a cosmologist's nightmare. There are numerous classes and subclasses of galaxies. Previously hundreds of thousands of volunteers helped classify millions of these images by eye. But with growing data, it becomes unfeasible to this manually. By using machine learning, we are automating the process of classifying galaxies using deep convolution neural network. The galaxies can be classified based on its features into 3 main categories: elliptical, Spiral and Irregular. By using machine learning we can reduce the human labor and the time required to complete such a herculean task. Currently this project only classifies for 3 broad classes and not for all the subclasses, so we are planning to expand this project to categorizing these classes of galaxies into their respective subclasses. We also hope to extend this project to include the classification of various other entities that make up our universe such as supernovae, pulsar stars, etc.

Things like growing volumes and varieties of available data, cheaper and more powerful computational processing, data storage and large-value predictions that can guide better decisions and smart actions in real time without human intervention are playing critical role in this age. All of these require models that can automatically analyse large complex data and deliver quick accurate results – even on a very large scale. Machine learning plays a significant role in developing these models. The applications of machine learning range from speech and object recognition to analysis and prediction of finance markets. Artificial Neural Network is one of the important algorithms of machine learning that is inspired by the structure and functional aspects of the biological neural networks. In this paper, we discuss the purpose, representation and classification methods for developing hardware for machine learning with the main focus on neural networks. This paper also presents the requirements, design issues and optimization techniques for building hardware architecture of neural networks. KEYWORDS Artificial intelligence (AI), application specific integrated circuit (ASIC), artificial neural network (ANN), central processing unit (CPU), field programmable gate array (FPGA), graphics processing unit (GPU), machine learning (ML), neurochip

Over the decades, water pollution has been a real threat to the living species. The real-time monitoring of drinking water is nothing less than a challenging task. This paper aims to design and develop a low-cost system for the real-time monitoring of water quality using the Internet of Things (IoT) and Machine Learning (ML). The physical and chemical parameters of water such as temperature, level, moisture, humidity and visibility are measured using respective sensors. ESP8266, the core controller is employed to process the measured values from the sensors. The data acquired from Sensors are sent to the Django server. Random Forest (RF) and K-Nearest Neighbours (KNN) algorithm are used in the analysis and prediction of water quality.

Machine learning is modern and highly sophisticated technological applications became a huge trend in the health care industry. It provides methods, techniques and tools that can help in solving diagnostic problems in a variety of medical domains e.g. prediction of disease progression, extraction of medical knowledge for outcome research, therapy planning and support, and for the overall patient management. In this paper, various machine learning algorithms has been used for developing efficient decision support that can be used in healthcare applications. Cancer has been the leading cause of death worldwide for many years. The objective of this study is to diagnose and predict the possibility of occurrence of three most common types of cancer including Lung Cancer, Breast Cancer and Prostate Cancer using various Machine Learning supervised algorithms like Logistic regression, Naive Bayes classification, Kernel support vector classification , Decision Tree Classifier, Random Forest Classifier and XG Boost Classifier.

In this paper we examine ensemble methods for regression that leverage or "boost" base regressors by iteratively calling them on modified samples. The most successful leveraging algorithm for classification is AdaBoost, an algorithm that requires only modest assumptions on the base learning method for its strong theoretical guarantees. We present several gradient descent leveraging algorithms for regression and prove AdaBoost-style bounds on their sample errors using intuitive assumptions on the base learners. We bound the complexity of the regression functions produced in order to derive PAC-style bounds on their generalization errors. Experiments validate our theoretical results.

Comparison study of algorithms is very much required before implementing them for the needs of any  organization. The comparisons of algorithms are depending on the various parameters such as data frequency, types of data and relationship among the attributes in a given data set. There are number of learning and classifications algorithms are used to analyse, learn patterns and categorize data are available. But the problem is the one to find the best algorithm according to the problem and desired output. The desired result has always been higher accuracy in predicting future values or events from the given dataset. Algorithms taken for the comparisons study are Neural net, SVM, Naïve Bayes, BFT and Decision stump. These top algorithms are most influential data mining algorithms in the research community. These algorithms have been considered and mostly used in the field of knowledge discovery and data mining.

The goal of approximate policy evaluation is to “best” represent a target value function according to a specific criterion. Different algorithms offer different choices of the optimization criterion. Two popular least-squares algorithms for performing this task are the Bellman residual method, which minimizes the Bellman residual, and the fixed point method, which minimizes the projection of the Bellman residual. When used within policy iteration, the fixed point algorithm tends to ultimately find better performing policies whereas the Bellman residual algorithm exhibits more stable behavior between rounds of policy iteration. We propose two hybrid least-squares algorithms to try to combine the advantages of these algorithms. We provide an analytical and geometric interpretation of hybrid algorithms and demonstrate their utility on a simple problem. Experimental results on both small and large domains suggest hybrid algorithms may find solutions that lead to better policies when performing policy iteration.

We consider two on-line learning frameworks: binary classification through linear threshold functions and linear regression. We study a family of on-line algorithms, called p-norm algorithms, introduced by Grove, Littlestone and Schuurmans in the context of deterministic binary classification. We show how to adapt these algorithms for use in the regression setting, and prove worst-case bounds on the square loss, using a technique from Kivinen and Warmuth. As pointed out by Grove, et al., these algorithms can be made to approach a version of the classification algorithm Winnow as p goes to infinity; similarly they can be made to approach the corresponding regression algorithm EG in the limit. Winnow and EG are notable for having loss bounds that grow only logarithmically in the dimension of the instance space. Here we describe another way to use the p-norm algorithms to achieve this logarithmic behavior. With the way to use them that we propose, it is less critical than with Winnow and EG to retune the parameters of the algorithm as the learning task changes. Since the correct setting of the parameters depends on characteristics of the learning task that are not typically known a priori by the learner, this gives the p-norm algorithms a desireable robustness. Our elaborations yield various new loss bounds in these on-line settings. Some of these bounds improve or generalize known results. Others are incomparable.

While many constructive induction algorithms focus on generating new binary attributes, this paper explores novel methods of constructing nominal and numeric attributes. We propose a new constructive operator, X-of-N. An X-of-N representation is a set containing one or more attribute-value pairs. For a given instance, the value of an X-of-N representation corresponds to the number of its attribute-value pairs that are true of the instance. A single X-of-N representation can directly and simply represent any concept that can be represented by a single conjunctive, a single disjunctive, or a single M-of-N representation commonly used for constructive induction, and the reverse is not true. In this paper, we describe a constructive decision tree learning algorithm, called XofN. When building decision trees, this algorithm creates one X-of-N representation, either as a nominal attribute or as a numeric attribute, at each decision node. The construction of X-of-N representations is carried out by greedily searching the space defined by all the attribute-value pairs of a domain. Experimental results reveal that constructing X-of-N attributes can significantly improve the performance of decision tree learning in both artificial and natural domains in terms of higher prediction accuracy and lower theory complexity. The results also show the performance advantages of constructing X-of-N attributes over constructing conjunctive, disjunctive, or M-of-N representations for decision tree learning.

An important application of reinforcement learning (RL) is to finite-state control problems and one of the most difficult problems in learning for control is balancing the exploration/exploitation tradeoff. Existing theoretical results for RL give very little guidance on reasonable ways to perform exploration. In this paper, we examine the convergence of single-step on-policy RL algorithms for control. On-policy algorithms cannot separate exploration from learning and therefore must confront the exploration problem directly. We ...

This paper describesfoil, a system that learns Horn clauses from data expressed as relations.foil is based on ideas that have proved effective in attribute-value learning systems, but extends them to a first-order formalism. This new system has been applied successfully to several tasks taken from the machine learning literature.

Backpropagation, similar to most learning algorithms that can form complex decision surfaces, is prone to overfitting. This work presents classification-based objec-tive functions, an approach to training artificial neural networks on classification problems. Classification-...

In this paper we address two symmetrical issues, the discovery of similarities among classification algorithms, and among datasets. Both on the basis of error measures, which we use to define the error correlation between two algorithms, and determine the relative performance of a list of algorithms. We use the first to discover similarities between learners, and both of them to discover similarities between datasets. The latter sketch maps on the dataset space. Regions within each map exhibit specific patterns of error correlation or relative performance. To acquire an understanding of the factors determining these regions we describe them using simple characteristics of the datasets. Descriptions of each region are given in terms of the distributions of dataset characteristics within it.