INDIAN STATISTICAL INSTITUTE DOCUMENTATION RESEARCH AND TRAINING CENTRE 8th Mile Mysore Road, Bangalore-560059 DRTC Seminar 2012-13 Pattern Recognition with Semi-Supervised Learning Algorithm Pattern Recognition principles makes the machines to take decision, just like human beings to sense the environment and take action according to what they observe. Taking decision for a given input based on the prior knowledge is the basis of a pattern recognition process. For example, to decide the category of a mail, such as “spam” or “non-spam” based on its contents. Pattern recognition is performed with three basic learning methodologies, such as supervised, unsupervised and Semi-Supervised. Supervised learning is the task of inferring a function from a labeled data consisting of a set of examples. A supervised learning algorithm analyzes the training data and produces an inferred function, which is called a “classifier” (if the o/p is discrete) or a regression function (if o/p is continuous). Unsupervised learning refers to the problem of trying to find hidden structure in unlabeled data. Semi-Supervised learning is a combination of supervised and unsupervised notions and in recent past, it has gained its momentum in the decision-making process. Supervised learning algorithms require enough labeled training data to learn reasonably accurate classifiers. Unsupervised learning methods are employed to discover structure in unlabeled data. Semi-supervised learning allows taking advantage of the strengths of both. The goal of semi-supervised learning is to understand how combining labeled and unlabeled data may change the learning behavior, and design algorithms that take advantage of such a combination. Many machine-learning algorithms have been designed, which revealed that using unlabeled data in combination with a small amount of labeled data could produce considerable improvement in learning accuracy. The acquisition of labeled data for a learning problem often requires a skilled human agent (e.g. to translate a medical observation) or a physical experiment (e.g. determining the 3D structure of a protein or determining whether there is oil at a particular location). In many practical learning domains, there is a large supply of unlabeled data but limited labeled data, which can be expensive to generate, e.g., text processing, video-indexing, bioinformatics. In such situations, semi-supervised learning can be of great practical value. Semi-supervised learning is also of theoretical interest in machine learning and as a model for human learning because it works with both labeled and unlabeled data set. All are Cordially Invited Speaker: Anurodh kumar Sinha Venue: DRTC Classroom Date: 29th Nov.12 Time: 2:00 pm Dr. Devika P. Madalli / Dr. Saiful Amin Seminar coordinator -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.