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

The measurement of meaning

Classic IR (information retrieval) is inherently predicated on users searching for information, the so-called "information need". But the need behind a web search is often not informational -- it might be navigational (give me the url of the site I want to reach) or transactional (show me sites where I can perform a certain transaction, e.g. shop, download a file, or find a map). We explore this taxonomy of web searches and discuss how global search engines evolved to deal with web-specific needs.

/pdf/a-taxonomy-of-web-search-1c1wfauugs.pdf

A taxonomy of web search

A comparative analysis of major online review platforms: Implications for social media analytics in hospitality and tourism

Over the recent years, the growth of online social media has greatly facilitated the way people communicate with each other. Users of online social media share information, connect with other people and stay informed about trending events. However, much recent information appearing on social media is dubious and, in some cases, intended to mislead. Such content is often called fake news. Large amounts of online fake news has the potential to cause serious problems in society. Many point to the 2016 U.S. presidential election campaign as having been influenced by fake news. Subsequent to this election, the term has entered the mainstream vernacular. Moreover it has drawn the attention of industry and academia, seeking to understand its origins, distribution and effects. Of critical interest is the ability to detect when online content is untrue and intended to mislead. This is technically challenging for several reasons. Using social media tools, content is easily generated and quickly spread, leading to a large volume of content to analyse. Online information is very diverse, covering a large number of subjects, which contributes complexity to this task. The truth and intent of any statement often cannot be assessed by computers alone, so efforts must depend on collaboration between humans and technology. For instance, some content that is deemed by experts of being false and intended to mislead are available. While these sources are in limited supply, they can form a basis for such a shared effort. In this survey, we present a comprehensive overview of the finding to date relating to fake news. We characterize the negative impact of online fake news, and the state-of-the-art in detection methods. Many of these rely on identifying features of the users, content, and context that indicate misinformation. We also study existing datasets that have been used for classifying fake news. Finally, we propose promising research directions for online fake news analysis.

An overview of online fake news: Characterization, detection, and discussion

Using human evaluation of 100,000 words spread across 24 corpora in 10 languages diverse in origin and culture, we present evidence of a deep imprint of human sociality in language, observing that (i) the words of natural human language possess a universal positivity bias, (ii) the estimated emotional content of words is consistent between languages under translation, and (iii) this positivity bias is strongly independent of frequency of word use. Alongside these general regularities, we describe interlanguage variations in the emotional spectrum of languages that allow us to rank corpora. We also show how our word evaluations can be used to construct physical-like instruments for both real-time and offline measurement of the emotional content of large-scale texts.

/pdf/human-language-reveals-a-universal-positivity-bias-21q4wsl04n.pdf

Human language reveals a universal positivity bias

This report describes the Shiraz project MT prototype for a Persian to English machine translation system using typed feature structures and unification. An overview of the linguistic properties of Persian is presented and the morphological and syntactic grammars developed within the Shiraz project are discussed. The underlying model for the system is a layered chart, capable of representing heterogeneous types of hypotheses in an integrated way.

Persian-English Machine Translation: An Overview of the Shiraz Project

Cet article etudie les proprietes syntaxiques et semantiques des verbes composes en persan afin de mettre a jour les contributions individuelles des composants de la construction verbale. La structure evenementielle de certains predicats complexes est determinee et une analyse de la construction verbale est proposee ou les composants sont decomposes en elements syntaxiques primitifs qui consistent en des racines lexicales et des tetes fonctionnelles, ces dernieres etant responsables de la projection des arguments de la construction verbale. L'analyse fournie suggere que la structure argumentale n'est pas projetee du lexique, mais est plutot formee de facon compositionnelle en joignant, dans le domaine syntaxique, les composants primitifs du predicat complexe. Les travaux portant sur les verbes composes en persan ont indique que ces constructions montrent un double comportement lexical et syntaxique. L'analyse proposee peut facilement expliquer ces proprietes puisqu'il n'y existe aucune division stricte entre le niveau responsable pour la formation des mots et le module qui manipule les constructions phrastiques.

Event Structure and Complex Predicates in Persian

This report provides a complete descriptive analysis of Persian inflectional morphology from a computational perspective. The parts of speech and the morphemes that appear on them as well as their corresponding morphotactics are presented in detail. The verbal paradigm is also described in this document. Since the morphological analyzer designed for this project uses a unification-based grammar with typed feature structures, the morphological information has been defined in terms of features and values. The report describes the current version of the morphological analyzer used in the Shiraz project and discusses any morphological elements that have not been included in this version, mostly due to the colloquial usage of these morphemes. Sample rules of Samba, the grammar specifying the morphological analyzer, as well as the feature specification for the Persian type definitions module are also described.

Persian Computational Morphology: A Unification-Based Approach

The nature of preverbal nominals and their relation to the verb have been the focus of much debate in languages with a productive complex predication process. For Persian, certain analyses have argued that the bare nominals in complex predicate constructions are distinct from bare objects, while others have treated the two types of bare nominals uniformly. This paper argues that the two categories of preverbal nouns cannot receive the same analysis since they display distinct syntactic and semantic behavior: the preverbal nominals, unlike the bare object nouns, cannot be questioned, are modified differently, have different interpretations, give rise to distinct case-assignment contexts, and can co-occur with a non-specific object. The distinct properties of the two nominal categories are captured by positing distinct structural positions for these nouns. Non-specific bare nouns are internal arguments of the thematic verb, while the nominal element of the complex predicate construction is part of the verbal domain with which it combines through a process of conflation, as defined in Hale and Keyser (2002), to form a single predicate.

Karine Megerdoomian

Papers

Human language reveals a universal positivity bias

Persian-English Machine Translation: An Overview of the Shiraz Project

Event Structure and Complex Predicates in Persian

Persian Computational Morphology: A Unification-Based Approach

The Status of the Nominal in Persian Complex Predicates