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

We will review some of the major results in random graphs and some of the more challenging open problems. We will cover algorithmic and structural questions. We will touch on newer models, including those related to the WWW.

Random graphs

Sentiment analysis and opinion mining is the field of study that analyzes people's opinions, sentiments, evaluations, attitudes, and emotions from written language. It is one of the most active research areas in natural language processing and is also widely studied in data mining, Web mining, and text mining. In fact, this research has spread outside of computer science to the management sciences and social sciences due to its importance to business and society as a whole. The growing importance of sentiment analysis coincides with the growth of social media such as reviews, forum discussions, blogs, micro-blogs, Twitter, and social networks. For the first time in human history, we now have a huge volume of opinionated data recorded in digital form for analysis. Sentiment analysis systems are being applied in almost every business and social domain because opinions are central to almost all human activities and are key influencers of our behaviors. Our beliefs and perceptions of reality, and the choices we make, are largely conditioned on how others see and evaluate the world. For this reason, when we need to make a decision we often seek out the opinions of others. This is true not only for individuals but also for organizations. This book is a comprehensive introductory and survey text. It covers all important topics and the latest developments in the field with over 400 references. It is suitable for students, researchers and practitioners who are interested in social media analysis in general and sentiment analysis in particular. Lecturers can readily use it in class for courses on natural language processing, social media analysis, text mining, and data mining. Lecture slides are also available online.

/pdf/sentiment-analysis-and-opinion-mining-8314pykwrf.pdf

Sentiment Analysis and Opinion Mining

/pdf/semi-supervised-learning-literature-survey-4ccxy3rmmj.pdf

Semi-Supervised Learning Literature Survey

In the field of machine learning, semi-supervised learning (SSL) occupies the middle ground, between supervised learning (in which all training examples are labeled) and unsupervised learning (in which no label data are given). Interest in SSL has increased in recent years, particularly because of application domains in which unlabeled data are plentiful, such as images, text, and bioinformatics. This first comprehensive overview of SSL presents state-of-the-art algorithms, a taxonomy of the field, selected applications, benchmark experiments, and perspectives on ongoing and future research. Semi-Supervised Learning first presents the key assumptions and ideas underlying the field: smoothness, cluster or low-density separation, manifold structure, and transduction. The core of the book is the presentation of SSL methods, organized according to algorithmic strategies. After an examination of generative models, the book describes algorithms that implement the low-density separation assumption, graph-based methods, and algorithms that perform two-step learning. The book then discusses SSL applications and offers guidelines for SSL practitioners by analyzing the results of extensive benchmark experiments. Finally, the book looks at interesting directions for SSL research. The book closes with a discussion of the relationship between semi-supervised learning and transduction. Adaptive Computation and Machine Learning series

/pdf/semi-supervised-learning-scg1exixez.pdf

Semi-Supervised Learning

Target detection in hyperspectral images (HSIs) is an active area of research; it seeks to detect objects that are small in both number and size within a scene. The proposed work presents a new methodology for target detection in HSIs by combining kurtosis, level sets, and a size-based thresholding strategy. Kurtosis is used as a preprocessing step to initially enhance the targets in an image. Then, level sets identify and mark associations of pixels with similar spectral information as candidate targets. Finally, the size-based thresholding strategy detects true targets and discards false alarms that do not fit with target dimensions set as input parameter. In addition, we propose a novel version of level sets, which is suitable for target detection tasks in HSIs. Results show that the proposed algorithm could successfully detect targets in HSIs, and it gave better performance in terms of the receiver operating characteristic curve than other techniques widely used in target detection such as orthogonal subspace projection, constrained signal detector, constrained energy minimization, adaptive cosine/coherent estimator algorithm, and generalized-likelihood ratio test.

A New Methodology Based on Level Sets for Target Detection in Hyperspectral Images

In this paper, we address the problem of delivering content in a pervasive environment characterized by high variability in network conditions, client devices and user context (e.g. location and preferences). This variability results in non-uniform user experience due to high and variable latency and could lead to user frustration during service access. We propose a methodology to tackle this issue and present a test-bed system that we are currently developing to verify our hypotheses.

/pdf/preparing-the-edge-of-the-network-for-pervasive-content-1fgqmd4juh.pdf

Preparing the Edge of the Network for Pervasive Content Delivery

Electron Cryo-microscopy is an advanced imaging technique that is able to produce volumetric images of proteins that are large or hard to crystallize. De novo modeling is a process that aims at deriving the structure of the protein using the images produced by Electron Cryo-microscopy. At the medium resolutions (5 to 10A), the location and orientation of the secondary structure elements can be computationally identified on the images. However, there is no registration between the detected secondary structure elements and the protein sequence, and therefore it is challenging to derive the atomic structure from such volume data. The skeleton of the volume image is used to interpret the connections between the secondary structure elements in order to reduce the search space of the registration problem. Unfortunately, not all features of the image can be captured using a single segmentation. Moreover, the skeleton is sensitive to the threshold used which leads to gaps in the skeleton. In this paper, we present a threshold-independent approach to overcome the problem of gaps in the skeletons. The approach uses a novel representation of the image where the image is modeled as a graph and a set of volume trees. A test containing thirteen synthesized images and two authentic images showed that our approach could improve the existent skeletons. The percent of improvement achieved were 117% and 40% for Gorgon and MapEM, respectively.

A Graph Approach to Bridge the Gaps in Volumetric Electron Cryo-microscopy Skeletons

In many machine learning application domains obtaining labeled data is expensive but obtaining unlabeled data is much cheaper. For this reason there has been growing interest in algorithms that are able to take advantage of unlabeled data. In this thesis we develop several methods for taking advantage of unlabeled data in classification and regression tasks. 
Specific contributions include: (1) A method for improving the performance of the graph mincut algorithm of Blum and Chawla [12] by taking randomized mincuts. We give theoretical motivation for this approach and we present empirical results showing that randomized mincut tends to outperform the original graph mincut algorithm, especially when the number of labeled examples is very small. (2) An algorithm for semi-supervised regression based on manifold regularization using local linear estimators. This is the first extension of local linear regression to the semi-supervised setting. In this thesis we present experimental results on both synthetic and real data and show that this method tends to perform better than methods which only utilize the labeled data. (3) An investigation of practical techniques for using the Winnow algorithm (which is not directly kernelizable) together with kernel functions and general similarity functions via unlabeled data. We expect such techniques to be particularly useful when we have a large feature space as well as additional similarity measures that we would like to use together with the original features. This method is also suited to situations where the best performing measure of similarity does not satisfy the properties of a kernel. We present some experiments on real and synthetic data to support this approach.

/pdf/techniques-for-exploiting-unlabeled-data-37irj8qkeg.pdf

Mugizi Robert Rwebangira

Papers

A New Methodology Based on Level Sets for Target Detection in Hyperspectral Images

Preparing the Edge of the Network for Pervasive Content Delivery

A Graph Approach to Bridge the Gaps in Volumetric Electron Cryo-microscopy Skeletons

Techniques for exploiting unlabeled data

Mono-isotope prediction for mass spectra using Bayes network