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

A morphable model for the synthesis of 3D faces

On the origin of the species

A web service is a modular application that is published, advertised, discovered, and invoked across a network, i.e., an intranet or the Internet. It is based on a software-as-services model and may participate as a component of other web services and applications. Binary and XML are two popular encoding/decoding mechanisms for network messages. Binary encoding is used when performance is critical and XML encoding is employed when interoperability with other web services and applications is essential. With each, one may employ compression to reduce message size prior to its transmission across the network. These decisions have a significant impact on response time and throughput. This paper reports on our experiences with a decision support benchmark, TPGH, using these alternatives on different hardware platforms. We focus on queries and make the following observations. First, compression reduces the message size and enhances the throughput of a shared network. With XML, we present numbers from XMill, a compression technique that employs XML semantics. For queries that produce more than one megabyte of XML data, XMill compressed XML messages are almost always smaller than the Zip compressed Binary messages. While this improves the throughput of a networked environment with a fixed bandwidth, the response time of XMill compressed messages are at times twice slower than Zip compressed Binary messages. The processor speed has a significant impact on the observed response times.

A comparison of alternative encoding mechanisms for web services

In this work, a spoken query system developed for accessing the price of agricultural commodities in Assamese language is described. The developed system enables the user to access the latest price of the commodity by calling the system using a landline/mobile phone. The spoken query system consists of interactive voice response (IVR) and automatic speech recognition (ASR) modules which are developed using open source resources. For the development of ASR models, the task specific speech data is collected from different dialect regions of Assam. The issues in test data adaptation are highlighted and a constrained data-unseen speaker adaptation approach is implemented which is found to give a relative improvement by 8% in baseline performance.

Assamese spoken query system to access the price of agricultural commodities

In this paper, we have prepared a medium size Bangla speech corpus and compare performances of different acoustic features for Bangla word recognition. Most of the Bangla automatic speech recognition (ASR) system uses a small number of speakers, but 40 speakers selected from a wide area of Bangladesh, where Bangla is used as a native language, are involved here. In the experiments, mel-frequency cepstral coefficients (MFCCs) are inputted to the triphone hidden Markov model (HMM) based classifiers for obtaining word recognition performance. From the experiments, it is shown that MFCC-based method of 39 dimensions provides a higher word correct rate (WCR) and word accuracy (WA) than the other methods investigated. Moreover, a higher WCR and WA is obtained by the MFCC39-based method with fewer mixture components in the HMM.

Bangla triphone HMM based word recognition

We proposed a classification technique of EEG motor imagery signals using Logistic regression and feature extraction algorithm using filter bank common spatial pattern (FBCSP). Main theme of FBCSP is that the signals decomposed into 5 sub band then calculated CSP for each sub band, this algorithm also allows automated frequency band selection. We combined each subband CSP feature vector, feed this feature vector into machine learning algorithm. In the paper Logistic regression is used to classify among multiple classes. To evaluate this method, we used here publicly available dataset namely Brain-Computer Interface competition IV-2a. Because of high accuracy and kappa that shown in accuracy table that proposed method is promising.

/pdf/multiclass-mi-task-classification-using-logistic-regression-v003n22qfo.pdf

Multiclass MI-Task Classification Using Logistic Regression and Filter Bank Common Spatial Patterns

Quick Sort is considered as the fastest sorting algorithm among all the sorting algorithms. The idea of selecting a pivot was introduced in classical Quick Sort in 1962. This sorting algorithm takes favorably less time compared to other methods. It needs a complex time O(\( nlogn \)) for the best case and O(\( n^{2} \)) for worst-case which occurs when the input array is already sorted or reversely sorted. To reduce the worst-case complexity we provide a strong algorithm where it makes fewer comparisons and the time complexity after using this algorithm becomes a function of logarithm O(\( nlogn \)) for worst-case complexity. We experimentally evaluate our algorithms and compare them with classical algorithms and with other papers. The algorithm presented here has profound implications for future studies of handling worst-case complexity and may one day help to solve this occurrence of the fastest sorting method.

Optimizing Complexity of Quick Sort

This paper describes an evaluation ofInhibition/Enhancement (In/En) network for robustautomatic speech recognition (ASR). In distinctive phoneticfeatures (DPFs) based speech recognition using neuralnetwork, In/En network is needed to discriminate whetherthe DPFs dynamic patterns of trajectories are convex orconcave. The network is used to achieve categorical DPFsmovement by enhancing DPFs peak patterns (convexpatterns) and inhibiting DPFs dip patterns (concavepatterns). We have analyzed the effectiveness of In/Enalgorithm by incorporating it into a system which consists ofthree stages: a) Multilayer Neural Networks (MLNs), b)In/En Network and c) Gram-Schmidt (GS)orthogonalization. From the experiments using JapaneseNewspaper Article Sentences (JNAS) database in clean andnoisy acoustic environments, it is observed that the In/Ennetwork plays a significant role on the improvement ofphoneme recognition performance. Moreover, In/Ennetwork reduces required number of mixture componentsin Hidden Markov Models (HMMs).

Inhibition/Enhancement Network Based ASR using Multiple DPF Extractors

The classical algorithm which is dedicated to resolve job sequencing problem with a deadline (JSD) needs exponential time O(\( n^{2} \)), where sorting algorithm [O(\( nlog\left( n \right) \))-(Merge Sort)] must have to use to sort all the jobs in decreasing order of their profit and it is a greedy technique. To reduce the complexity of this classical algorithm, we nullify the sorting algorithm using dynamic programming approach in the proposed algorithm. The time complexity after using this approach reduces to O(\( mn \)), where no sorting algorithm [O(\( nlog\left( n \right) \))-(Merge Sort)] needed, which has been shown by proper explanation. Here, we were also given a novel approach to resolve the job sequencing problem using Dynamic Programming and it is a unique approach that always finds an optimal solution. By using this approach, a proper algorithm has been developed in this paper. Besides, finding maximum profit and the sequence of the job to obtain maximum profit, this algorithm gives the sequence of jobs for a specific profit or near a specific profit.

Mohammad Hasan

Papers

Bangla triphone HMM based word recognition

Multiclass MI-Task Classification Using Logistic Regression and Filter Bank Common Spatial Patterns

Optimizing Complexity of Quick Sort

Inhibition/Enhancement Network Based ASR using Multiple DPF Extractors

A New Approach to Solve Job Sequencing Problem Using Dynamic Programming with Reduced Time Complexity