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

Emotion and Adaptation

A field is emerging that leverages the capacity to collect and analyze data at a scale that may reveal patterns of individual and group behaviors.

/pdf/computational-social-science-5el7814wob.pdf

Computational Social Science

This survey covers fifteen years of research in the Named Entity Recognition and Classification (NERC) field, from 1991 to 2006. We report observations about languages, named entity types, domains and textual genres studied in the literature. From the start, NERC systems have been developed using hand-made rules, but now machine learning techniques are widely used. These techniques are surveyed along with other critical aspects of NERC such as features and evaluation methods. Features are word-level, dictionary-level and corpus-level representations of words in a document. Evaluation techniques, ranging from intuitive exact match to very complex matching techniques with adjustable cost of errors, are an indisputable key to progress.

A survey of named entity recognition and classification

which is used to describe a third component of thinking processes, particularly preverbal, and it denotes the concept that the world is activated by some generalized "energy" that links together causally all objects and events ; it is presumably revealed by a person's lack of curiosity about causal connections, as though they were self-evident. Aside from the rather frequent use of such key words, having strong connotations for this reviewer far away from what the author is aiming to denote, the book is written in a lucid and stimulating style that makes reading it an invigorating intellectual exercise. It is a book that is likely to have somewhat limited attraction to the full-time clinician, especially one treating adult patients. And child psychiatrists and psychologists, if reasonably well read, will most likely be familiar with the majority of references from which this author has synthesized his material. On the other hand, the scholarly and refreshing con¬ ceptual approaches of the author will appeal to psychologists, philosophers, linguists, and psychiatrists with a research bent and anyone else who wants to be provoked to do some thinking on the problems of language, language development, and the psychology of cognition. Louis A. GOTTSCHALK, M.D. Plans and the Structure of Behavior. By George A Miller, Eugene gALANTER, and Karl H. Pribram. Price, $5.00. Pp. 226. Henry Holt & Co., Inc., New York 17, 1960. This is an important book for psychiatrists and behavioral scientists, since it presents a clear, concise study of the application of cybernetics, information and computer theory to the problem of analyzing behavior. The authors have been actively engaged in behavioral research in different areas\p=m-\Millerin information and communication, Galanter in experimental psychology, and Pribram in neurophysiology. The book resulted from a series of discussions which they engaged in during a year they spent together at the Center for Advanced Study in the Behavioral Sciences, Palo Alto, Calif. Their original intent was to write a diary, as it were, of the development of their ideas and, fortunately, enough of this remains to make the book clear, easy to read, and interesting. It is also fortunate, however, that in the final writing a variety of studies comparing the "behavior" of computing machines with human "cognitive behavior" have been reviewed and summarized. The result is one of the best presentations of the present status of the brain-computer problem. The authors, however, do not discuss certain aspects of this problem, such as the difference between the brain and its vast number of parallel channels, but few operations, and the modern high-speed computer with its few channels and vast numbers of operations. This omission is consistent with their interest, since it would introduce the question of mechanisms rather than the problem of the structure of behavior as it is observed in everyday life in the clinic and in experiments on learning, conditioning, etc. Similarly, they do not discuss the qualitative differences between mechanisms of memory in the computer and those in the brain. In the former, a "memory"\p=m-\ i.e., stored information\p=m-\isidentified, metaphorically speaking, by an address, whereas no such mechanism is known in the brain (personal communication, Dr. Julian Bigelow). With few exceptions, however, the data, concepts, and theories presented are handled with elegant precision, as illustrated by the discussion of Sherrington's concepts of the "Reflex" and the "Synapse," Kurt Lewin's ideas of "tension states," and the numerous references to the work of Newell, Shaw, and Simon on computers and logic. There are, nevertheless, areas with

https://link.springer.com/content/pdf/10.1057%2Fjors.1968.86.pdf

Plans and the Structure of Behavior

While Named Entity extraction is useful in many natural language applications, the coarse categories that most NE extractors work with prove insufficient for complex applications such as Question Answering and Ontology generation. We examine one coarse category of named entities, persons, and describe a method for automatically classifying person instances into eight finer-grained subcategories. We present a supervised learning method that considers the local context surrounding the entity as well as more global semantic information derived from topic signatures and WordNet. We reinforce this method with an algorithm that takes advantage of the presence of entities in multiple contexts.

/pdf/fine-grained-classification-of-named-entities-2xfimp4wm9.pdf

Fine grained classification of named entities

Social media content items are mapped to relevant time-based media events. These mappings are used as the basis for calculating metrics based upon the mappings, and ratings of the time-based media there from. Audience response metrics (ARMs) are calculated from the mappings, which provide an indication of audience engagement or response. In addition, ARMs provide information about the virality, depth, and breadth of the response by the viewing audience. The calculated metrics are stored and are used to generate ratings for the time-based media. The ratings may be published in whole or in part, e.g., as part of a ratings system.

Generating audience response metrics and ratings from social interest in time-based media

https://www.microsoft.com/en-us/research/wp-content/uploads/2006/06/Cogsci06_HumanSpeechome.pdf

The Human Speechome Project

Social media content items are mapped to relevant time-based media events. These mappings may be used as the basis for multiple applications, such as ranking of search results for time-based media, automatic recommendations for time-based media, prediction of audience interest for media purchasing/planning, and estimating social interest in the time-based media. Social interest in time-based media (e.g., video and audio streams and recordings) segments is estimated through a process of data ingestion and integration. The estimation process determines social interest in specific events represented as segments in time-based media, such as particular plays in a sporting event, scenes in a television show, or advertisements in an advertising block. The resulting estimates of social interest also can be graphically displayed.

Michael Fleischman

Papers

Fine grained classification of named entities

Generating audience response metrics and ratings from social interest in time-based media

The Human Speechome Project

Estimating social interest in time-based media

Displaying Estimated Social Interest in Time-based Media