Machine Learning - Learning Techniques, CNN, Languages and APIs

Abstract

Nowadays, Artificial intelligence is an important part in everyone's life. It can be derived in two categories named as Machine learning and deep learning. Machine learning is the emerging field of the current era. With the help of the machine learning, we can develop the computers in such a way so that they can learn themselves. There are various types of leaning algorithms used for machine learning. With the help of these algorithms, machines can learn various things and they can behave almost like the human beings. Nowadays, the role of the machine is not limited in some defined fields only; it is playing an important role in almost every field such as education, entertainment, medical diagnosis etc. In this research paper, the basics about machine learning is discussed we have discussed about various learning techniques such as supervised learning, unsupervised learning and reinforcement learning in detail. A small portion is also used to cover some basics about the Convolutional Neural Networks (CNN). Some information about the various languages and APIs, designed and mostly used for Machine Learning and its applications are also provided in this paper.

A comparative analysis of haemoglobin variants using machine learning algorithms

“…Also, from Figure 1, it is shown that the haemoglobin genotype variants used in this study are AA, AC, AS, SC, SS. The haemoglobin genotype variants were obtained from the analytical records of the Electrophoretic method using Helena Electrophoretic Machine [14], which has also been documented and cited in [15]- [20]. The recorded values of the haematological parameters were obtained from the automated analysis using Sysmex 5-part differential hematology analyzer.…”

A Novel Machine Learning Framework for Tracing Covid Contact Details by Using Time Series Locational data & Prediction Techniques

“…These artificial neural networks are model upon these neural networks in our brain, in ways like, these artificial neural networks are very interlinked with each other and rapidly transfer information between each other and there are also several layers of these neurons and the information or data is passed through all these layers and there will be a final output layer where we will be able to see the output of all the computation that has been done. These neural networks can then be tweaked little by little, to make them produce the desired result and therefore be trained properly [10]. This tweaking is done assigning weights to the nodes of the network and thereby affecting the data that is being passed through that node and with these tweaking of weights, the output is also changed and we can obtain our desired output by tweaking our way through these networks.…”