Parminder Singh Sethi 🇮🇳

Drug discovery is a time-consuming process, and it takes 10 to 15 years to develop a drug and obtain approval for the drug for public usage. Regardless of the huge investment and time commitment, not all drugs discovered using traditional methods are approved for public usage.
Because of the limitations of traditional drug discovery methods, computation models are being explored to accelerate the drug discovery process. In this paper, we explore the drawbacks of the traditional drug… Show more

Drug discovery is a time-consuming process, and it takes 10 to 15 years to develop a drug and obtain approval for the drug for public usage. Regardless of the huge investment and time commitment, not all drugs discovered using traditional methods are approved for public usage.
Because of the limitations of traditional drug discovery methods, computation models are being explored to accelerate the drug discovery process. In this paper, we explore the drawbacks of the traditional drug discovery method and how computation models such as machine learning models are used to accelerate the drug discovery process.
Furthermore, the paper explores the application of one of the machine learning frameworks that uses reinforcement learning to identify new drugs. The machine-learning framework consists of two models. The generative model builds rules to generate new molecules by using SMILES. The predictor model is used to predict the affinity of the newly generated molecules for the desired target, resulting in molecules with optimized desired properties. Show less

Our approach in this paper offers the application of Optical Character Recognition for the digitalization of handwritten medical prescriptions. It is determined that Doctor’s handwritten prescriptions create a two-fold problem, one of the problems is unreadable writing and another inability of pharmacists to understand the medicine name.
Our study will help in making pharmacist life easy at one end and making digitalized medical records maintenance easy. The field of OCR is still in the… Show more

Our approach in this paper offers the application of Optical Character Recognition for the digitalization of handwritten medical prescriptions. It is determined that Doctor’s handwritten prescriptions create a two-fold problem, one of the problems is unreadable writing and another inability of pharmacists to understand the medicine name.
Our study will help in making pharmacist life easy at one end and making digitalized medical records maintenance easy. The field of OCR is still in the evolution phase. OCR is a branch of computer vision in computer science.
Even after a lot of studies & related papers mentioning the procedures for translating handwritten textual content into computer-legible form, still it requires more research. Optical Character Recognition is a development process where the tools automatically recognize the image of any handwritten text/photo/scanned doc and transfer the same into character classification.
This research can be used as a guide and update for readers working in the Character Recognition area. Identification of a relevant feature mining process might be the single most valuable factor in attaining high character acknowledgment. Concise accuracy in character recognition processes without any disparity is one of the important measures for the success of this research.

OCR has been a subject of research for a long. As per our study, handwriting recognition is a classification problem. Optical Character Recognition (OCR) is a field of research in pattern recognition using artificial intelligence (AI). In OCR usually, the system scans and recognizes stationary images of the characters. It refers to the mechanical or electronic translation of images of handwritten characters into pixel-based machine code. Show less

In today’s world, traffic congestion has become a major issue in nearly all of the major cities around the globe. Due to an increase in the number of vehicles on the road, the intersections (or junctions) are major hotspots of congestion, so to maintain a proper flow, traffic intersections are installed with traffic lights that control the flow of the traffic moving in a particular direction. Traffic lights were first used nearly a century ago, but there have not been any major advancements in… Show more

In today’s world, traffic congestion has become a major issue in nearly all of the major cities around the globe. Due to an increase in the number of vehicles on the road, the intersections (or junctions) are major hotspots of congestion, so to maintain a proper flow, traffic intersections are installed with traffic lights that control the flow of the traffic moving in a particular direction. Traffic lights were first used nearly a century ago, but there have not been any major advancements in the functioning of traffic lights that can cope up with the requirements of today’s traffic. Nowadays, due to a lack of optimization in the timings of traffic lights, they can further cause more delays and congestions which causes environmental issues (like increased CO2 emissions and noise pollution), road rage, mental frustration, etc. This paper addresses the problem of traffic lights not having the optimized timings of signal for each direction due to the unavailability of robust, pro-active, and real-time analysis of traffic and then analyzing and predicting the best timings of traffic lights based upon the size (small, medium, or large) and class (emergency and normal) of vehicles. Show less

In today’s scenario, everybody wants to invest their money wisely to get high returns. Unfortunately, a plan that gives you high returns with low risks does not exist. However, to minimize the overall risk and to maximize the chance of high returns, it is recommended to diversify your investment portfolio in such a way that investment is done in multiple asset classes and securities. Diversification does not always help because it ensures low risk but cannot predict returns. Nowadays, machine… Show more

In today’s scenario, everybody wants to invest their money wisely to get high returns. Unfortunately, a plan that gives you high returns with low risks does not exist. However, to minimize the overall risk and to maximize the chance of high returns, it is recommended to diversify your investment portfolio in such a way that investment is done in multiple asset classes and securities. Diversification does not always help because it ensures low risk but cannot predict returns. Nowadays, machine learning is widely used in stock trading to predict and recommend actions to buy, sell, or hold. This is mostly restricted to a particular asset type such as equity. In general, we are likely to invest in different types of investment options, such as fixed deposits, mutual funds, equity, house, and so on. Therefore, it is wise to build a machine learning model which can recommend an investment portfolio across all investment options for an investor based on his/her expected returns and risk level. In this paper, we propose a machine learning model based on reinforcement learning that can continuously learn the emerging trend of each investment type and recommend the reallocation of assets to increase return on investment (ROI). Show less

With the advanced development in the Internet of Everything and related domains, tons of data is generated daily from sensors and devices. This data travels through different layers in a complex and distributed network, which opens the door for leakage and manipulation of data, exposing it to privacy breaches.

Fog computing or fog-networking, conjointly referred to as fogging, is an associate design that uses edge devices to hold out a considerable quantity of computation, storage, and… Show more

With the advanced development in the Internet of Everything and related domains, tons of data is generated daily from sensors and devices. This data travels through different layers in a complex and distributed network, which opens the door for leakage and manipulation of data, exposing it to privacy breaches.

Fog computing or fog-networking, conjointly referred to as fogging, is an associate design that uses edge devices to hold out a considerable quantity of computation, storage, and communication locally and routed over the Internet backbone. Also, fog computing is also known as the extension of the cloud computing paradigm to the edge of the network, therefore allowing an ecosystem to the various enterprise applications and services. The fog-computing services offer location sensitivity, decreased latency, geographical accessibility, connectivity (wireless), and improved data streaming.

In a fog network, various devices/nodes are added onto the trust factor among themselves as well as the fog node layer, which requires a good authentication method to establish this trustworthy relation/network among the different layers. Also, this trust/authentication is not only limited between nodes, but at the same time, fog-nodes communicate to cloud computing (vendors backend) and service the IoT devices/edge devices.

Additionally, trust and authentication are not the only challenges that exist today; there are more-like DOS attacks, end-user privacy, and many more. Therefore, maintaining privacy and providing secure channels is a huge and complicated task.

As part of this chapter, we will discuss the various challenges in existing robust and enterprise-level fog frameworks in the domain area of security and privacy. Show less

Cloud computing has emerged as an outstanding service delivery model. Almost all the organizations are switching to cloud in order to avoid the infrastructural cost and utilize the other benefits of the cloud services. As we know with benefits there come the issues, with tremendous increase in growth of technology, huge amount of data is being stored in cloud. Storage
of data in cloud requires a high cost of adaptability in order to provide an efficient, reliable and fault tolerance data… Show more

Cloud computing has emerged as an outstanding service delivery model. Almost all the organizations are switching to cloud in order to avoid the infrastructural cost and utilize the other benefits of the cloud services. As we know with benefits there come the issues, with tremendous increase in growth of technology, huge amount of data is being stored in cloud. Storage
of data in cloud requires a high cost of adaptability in order to provide an efficient, reliable and fault tolerance data storage system. Currently fault tolerance is handled by service provider or customer, which is an ineffective way of handling the fault tolerance and lead to in efficient solution. Rather than handling the problem over the service provider or customer this paper deals with an effective and innovative way of handling the fault tolerance in cloud. In this paper we propose a high level
approach by deploying the RAID-60 technology to service the fault tolerance problem in cloud. Show less

This paper introduces the concept of using IoT and Cloud based technology in car parking services in cities. A high-level view of the proposed system is outlined. Our solution makes the ancient parking system smarter by leveraging the power of IoT and embedding it with the latest innovation of electronic sensors & computers. An IoT-based intelligent car parking system is described. A number of software solutions, including Python, PHP web gateway with MySQL database, Cloud based storage and… Show more

This paper introduces the concept of using IoT and Cloud based technology in car parking services in cities. A high-level view of the proposed system is outlined. Our solution makes the ancient parking system smarter by leveraging the power of IoT and embedding it with the latest innovation of electronic sensors & computers. An IoT-based intelligent car parking system is described. A number of software solutions, including Python, PHP web gateway with MySQL database, Cloud based storage and mobile applications, are proposed to provide pleasant parking experience to mobile users. Also, Data generated by the sensors, Image detection cameras and mobile application will be used to gain insights by storing it in cloud foundry and applying Big Data analytics using Hadoop. Show less

Aim is to design a new way to provide the new and smart advertisements platform to all the 3 parties.
Basically in this project I want to reuse the channels that remained unused at the time of call connection.Through this project we are going to decrease the call rates. This model was represented in IIT-R .

Technologies : Networking, RR Protocol, Python (base language) and SDN(software defined networking).