Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

/pdf/a-break-in-the-clouds-towards-a-cloud-definition-svovsw1r5a.pdf

A Break in the Clouds: Towards a Cloud Definition

The Grid 2: Blueprint for a New Computing Infrastructure

The Internet of Things (IoT) is defined as a paradigm in which objects equipped with sensors, actuators, and processors communicate with each other to serve a meaningful purpose. In this paper, we survey state-of-the-art methods, protocols, and applications in this new emerging area. This survey paper proposes a novel taxonomy for IoT technologies, highlights some of the most important technologies, and profiles some applications that have the potential to make a striking difference in human life, especially for the differently abled and the elderly. As compared to similar survey papers in the area, this paper is far more comprehensive in its coverage and exhaustively covers most major technologies spanning from sensors to applications.

/pdf/internet-of-things-architectures-protocols-and-applications-46rpfbd58i.pdf

Internet of Things: Architectures, Protocols, and Applications

Cloud computing environments allow customers to dynamically scale their applications. The key problem is how to lease the right amount of resources, on a pay-as-you-go basis. Application re-dimensioning can be implemented effortlessly, adapting the resources assigned to the application to the incoming user demand. However, the identification of the right amount of resources to lease in order to meet the required Service Level Agreement, while keeping the overall cost low, is not an easy task. Many techniques have been proposed for automating application scaling. We propose a classification of these techniques into five main categories: static threshold-based rules, control theory, reinforcement learning, queuing theory and time series analysis. Then we use this classification to carry out a literature review of proposals for auto-scaling in the cloud.

A Review of Auto-scaling Techniques for Elastic Applications in Cloud Environments

The use of cloud computing and mobile devices is increasing in healthcare service delivery primarily because of the huge storage capacity of cloud, the heterogeneous structure of health data and the user-friendly interfaces on mobile devices. We propose a healthcare delivery scheme where a large knowledge base is stored in the cloud and user responses from mobile devices are input to this knowledge base to reach a preliminary diagnosis of diseases based on patients’ symptoms. However, instead of sending every response to the cloud and getting data from cloud server, it may often be desirable to prune a portion of the knowledge base that is stored in a graph form and download in to the mobile devices. Downloading data from cloud depends on the storage, battery power, processor of a mobile device, wireless network bandwidth and cloud processor capacity. In this paper, we focus on developing mathematical expressions involving the above mentioned criteria and show how these parameters are dependent on each other. The expressions built in this paper can be used in real-life scenarios to take decisions regarding the amount of data to be pruned in order to save energy as well as time.

Pruning of Health Data in Mobile-Assisted Remote Healthcare Service Delivery

Many dynamic programming approaches are existing for 1-0 Knapsack problem (KP) for fast GPU-based solution. These dynamic programming methods can be used for solving the problem of Bounded Knapsack Problem (BKP) after converting it into an equivalent 1-0 Knapsack problem. But, after conversion, large problem of BKP becomes too large 1-0 KP to fathoming via these methods on GPU-based system because of limited space availability in GPU device. In this paper, we present an implementation of an efficient dynamic programming technique to solve BKP on GPU-based system via CUDA. With an aspect to preserving the original problem size, BKP is not required to convert into equivalent 1/0 knapsack problem in this approach. Large problems are carried out with only minimally requisite CPU-GPU interactions and less memory occupancy on GPU for better efficiency and memory utilization. This GPU oriented parallel procedure shows brilliant speedup over sequential implementation of the dynamic technique.

An Efficient Reduced-Memory GPU-based Dynamic Programming Strategy for Bounded Knapsack Problems

For the dynamic nature of the grid, the applications need to adapt to the changing resource status and usage scenarios. This is required to ensure that the performances of the components do not suffer. Objective of this paper is to present an adaptive execution scheme for achieving guaranteed performance as has been agreed upon through service level agreements (SLAs). The scheme has been implemented based on the notion of performance properties and by deploying a set of autonomous agents within an integrated performance-based resource management framework.

Achieving guaranteed performance on computational Grid

The evolution of distributed architectures and programming paradigms for performance-oriented program development, challenge the state-of-the-art technology for performance tools. The area of high performance computing is rapidly expanding from single parallel systems to clusters and grids of heterogeneous sequential and parallel systems. Performance analysis and tuning applications is becoming crucial because it is hardly possible to otherwise achieve the optimum performance of any application. The objective of this paper is to study the state-of-the-art technology of the existing performance tools for distributed systems. The paper surveys some representative tools from different aspects in order to highlight the approaches and technologies used by them.

A Study on Performance Analysis Tools for Applications Running on Large Distributed Systems

In the present scenario, mobile communication is being widely used in the areas where infrastructure-based telecommunication systems are unavailable or to deploy one is not cost effective. However, often users suffer due to the fact that mobile devices have very low computational capability and thus are incapable of executing many applications with higher resource requirements. In a Mobile Ad-hoc Network (MANET) environment, when a node (a mobile device) is unable to accomplish a job, the job may be transferred to other nodes, which can fulfill its requirements in terms of resources. In ad-hoc network-based environments, such nodes should be found on demand, i.e. an algorithm should reach every node and match the requirements of a job with the capability of each node. This paper particularly emphasizes on the potentials of using PDAs or laptops of users within an office or in a meeting place. Users can submit their job requirements and then the request will pass through the MANET in order to search for a suitable node that would be able to respond back to the user. This paper proposes a Capability Searching Algorithm (CSA) for using in such scenario. CSA gathers information regarding the capability of a node during route discovery and users later can use this information in selecting the node and the best route to that node.

Nandini Mukherjee

Papers

Pruning of Health Data in Mobile-Assisted Remote Healthcare Service Delivery

An Efficient Reduced-Memory GPU-based Dynamic Programming Strategy for Bounded Knapsack Problems

Achieving guaranteed performance on computational Grid

A Study on Performance Analysis Tools for Applications Running on Large Distributed Systems

Capability aware AODV algorithm in ns-2 environment