Machine Learning is modern and highly sophisticated technological applications became a huge trend in the industry. Machine Learning is Omni present and is widely used in various applications. It is playing a vital role in many fields like finance, Medical science and in security. Machine learning is used to discover patterns from medical data sources and provide excellent capabilities to predict diseases. In this paper, we review various machine learning algorithms used for developing efficient decision support for healthcare applications. This paper helps in reducing the research gap for building efficient decision support system for medical applications.

Technological advances in computing and communications have lead to emerging technology called Cyber Physical Systems (CPS). CPS is the integration of the physical world with communication and computation capability. CPS are designed to achieve performance, robustness, reliability and efficiency in dealing with physical world. Many objects used in modern daily life are CPS.CPS will change the way people interface with the systems. This paper reviews an overview of CPS, security issues with CPS, challenges posed by CPS and some solutions for effective development and use of CPS pertaining to heath care domain. More research is needed to design and develop reliable and effective CPS.

A critical review of the Miller compensation technique for a two-stage operational amplifier (op-amp) is presented in this paper. The trade-offs involved in the compensation capacitor value and the small signal parameters of the op-amp are also considered in this paper; that is, the second stage requires a higher bias current, while driving a large capacitive load for a given phase-margin. A technique is presented with a view to increase the g m of the second stage without increasing the power dissipation, hence improving the phase margin while maintaining the unity gain bandwidth (UGB) of the op-amp. A prototype has been designed in 65nm CMOS technology and post-layout simulations exhibit good performance characteristics of 45MHz UGB with 250fF compensation capacitor, for a supply voltage of 1.2V, and a current consumption of $132\boldsymbol{\mu} \mathbf{A}$ . The new technique shows a superior performance to those obtained with an established compensation technique that achieves a 10MHz UGB while dissipating the same power. The total circuit size is 0.00234mm2 silicon area.

The industries and factories that produce goods which are always in demand require good amount of productivity, with small delays even can result in supply crunches. Most of the factories do not have smart machinery which can predict the breakdowns of the machines used and save time for better productivity because of the cost they come with. With a Predictive Maintenance system using machine learning on the edge this can save both time and cost of the factories and at the same time give better and quick results compared to cloud systems. In this paper we discuss on how the machine learning models can fit in a micro controller and run inference on it and predict results without requiring it to connect to a cloud system. We have emulated a washing machine to be a factory equipment and ran the system in it with its monitored data for predictions.

The growth of internet and network technologies has been increasing day by day.With the increase of these technologies, attacks and intrusions are also increasing. The prevention of these attacks has become an task. Intrusion Detection System (IDS) provides prevention against these attacks. Data Mining and Machine Learning techniques are used for IDS to reduce error rate and to improve accuracy and detection rate. In this paper, we proposed a novel Artificial Neural Network (ANN) classifier using Back propagation algorithm to model IDS. ANN is widely used supervised classifier for IDS. The performance of our model is evaluated by conducting experiments on KYOTO data set which is refined version of KDD99 data set. Empirical results show that proposed model is efficient with high detection rate and accuracy.

Cloud Computing is an emerging technology now a days , where cloud is most preferable when there is data backup, storage and data distribution service with low cost. But cloud is semi honest in nature due to not reveled storage and security structure thus while storing and sharing cloud data , its suppose to honest and secured. When data owners outsource their data in secured manner system should assure the security, data integrity and confidentiality. Here we noticed secure access controlling is the prime objective for sensitive data management.in this regards we have been used ABE (Attribute based encryption) for the above objective. This paper concentrates on secure storage and sharing system.