Opportunity and Curiosity find similar rocks on Mars. One can generally understand this statement if one knows that Opportunity and Curiosity are instances of the class of Mars rovers, and recognizes that, as signalled by the word on, rocks are located on Mars. Two mental operations contribute to understanding: recognize how entities/concepts mentioned in a text interact and recall already known facts (which often themselves consist of relations between entities/concepts). Concept interactions one identifies in the text can be added to the repository of known facts, and aid the processing of future texts. The amassed knowledge can assist many advanced language-processing tasks, including summarization, question answering and machine translation. Semantic relations are the connections we perceive between things which interact. The book explores two, now intertwined, threads in semantic relations: how they are expressed in texts and what role they play in knowledge repositories. A historical perspective takes us back more than 2000 years to their beginnings, and then to developments much closer to our time: various attempts at producing lists of semantic relations, necessary and sufficient to express the interaction between entities/concepts. A look at relations outside context, then in general texts, and then in texts in specialized domains, has gradually brought new insights, and led to essential adjustments in how the relations are seen. At the same time, datasets which encompass these phenomena have become available. They started small, then grew somewhat, then became truly large. The large resources are inevitably noisy because they are constructed automatically. The available corpora—to be analyzed, or used to gather relational evidence—have also grown, and some systems now operate at the Web scale. The learning of semantic relations has proceeded in parallel, in adherence to supervised, unsupervised or distantly supervised paradigms. Detailed analyses of annotated datasets in supervised learning have granted insights useful in developing unsupervised and distantly supervised methods. These in turn have contributed to the understanding of what relations are and how to find them, and that has led to methods scalable to Web-sized textual data. The size and redundancy of information in very large corpora, which at first seemed problematic, have been harnessed to improve the process of relation extraction/learning. The newest technology, deep learning, supplies innovative and surprising solutions to a variety of problems in relation learning. This book aims to paint a big picture and to offer interesting details.

People make sense of a text by identifying the semantic relations which connect the entities or concepts described by that text. A system which aspires to human-like performance must also be equipped to identify, and learn from, semantic relations in the texts it processes. Understanding even a simple sentence such as "Opportunity and Curiosity find similar rocks on Mars" requires recognizing relations (rocks are located on Mars, signalled by the word on) and drawing on already known relations (Opportunity and Curiosity are instances of the class of Mars rovers). A language-understanding system should be able to find such relations in documents and progressively build a knowledge base or even an ontology. Resources of this kind assist continuous learning and other advanced language-processing tasks such as text summarization, question answering and machine translation. The book discusses the recognition in text of semantic relations which capture interactions between base noun phrases. After a brief historical background, we introduce a range of relation inventories of varying granularity, which have been proposed by computational linguists. There is also variation in the scale at which systems operate, from snippets all the way to the whole Web, and in the techniques of recognizing relations in texts, from full supervision through weak or distant supervision to self-supervised or completely unsupervised methods. A discussion of supervised learning covers available datasets, feature sets which describe relation instances, and successful algorithms. An overview of weakly supervised and unsupervised learning zooms in on the acquisition of relations from large corpora with hardly any annotated data. We show how bootstrapping from seed examples or patterns scales up to very large text collections on the Web. We also present machine learning techniques in which data redundancy and variability lead to fast and reliable relation extraction.

Opportunity and Curiosity find similar rocks on Mars. One can generally understand this statement if one knows that Opportunity and Curiosity are instances of the class of Mars rovers, and recognizes that, as signalled by the word on, ROCKS are located on Mars. Two mental operations contribute to understanding: recognize how entities/concepts mentioned in a text interact and recall already known facts (which often themselves consist of relations between entities/concepts). Concept interactions one identifies in the text can be added to the repository of known facts, and aid the processing of future texts. The amassed knowledge can assist many advanced language-processing tasks, including summarization, question answering and machine translation. Semantic relations are the connections we perceive between things which interact. The book explores two, now intertwined, threads in semantic relations: how they are expressed in texts and what role they play in knowledge repositories. A historical perspective takes us back more than 2000 years to their beginnings, and then to developments much closer to our time: various attempts at producing lists of semantic relations, necessary and sufficient to express the interaction between entities/concepts. A look at relations outside context, then in general texts, and then in texts in specialized domains, has gradually brought new insights, and led to essential adjustments in how the relations are seen. At the same time, datasets which encompass these phenomena have become available. They started small, then grew somewhat, then became truly large. The large resources are inevitably noisy because they are constructed automatically. The available corpora—to be analyzed, or used to gather relational evidence—have also grown, and some systems now operate at the Web scale. The learning of semantic relations has proceeded in parallel, in adherence to supervised, unsupervised or distantly supervised paradigms. Detailed analyses of annotated datasets in supervised learning have granted insights useful in developing unsupervised and distantly supervised methods. These in turn have contributed to the understanding of what relations are and how to find them, and that has led to methods scalable to Web-sized textual data. The size and redundancy of information in very large corpora, which at first seemed problematic, have been harnessed to improve the process of relation extraction/learning. The newest technology, deep learning, supplies innovative and surprising solutions to a variety of problems in relation learning. This book aims to paint a big picture and to offer interesting details.

This book discusses the state of the art of automated essay scoring, its challenges and its potential. One of the earliest applications of artificial intelligence to language data (along with machine translation and speech recognition), automated essay scoring has evolved to become both a revenue-generating industry and a vast field of research, with many subfields and connections to other NLP tasks. In this book, we review the developments in this field against the backdrop of Elias Page's seminal 1966 paper titled "The Imminence of Grading Essays by Computer." Part 1 establishes what automated essay scoring is about, why it exists, where the technology stands, and what are some of the main issues. In Part 2, the book presents guided exercises to illustrate how one would go about building and evaluating a simple automated scoring system, while Part 3 offers readers a survey of the literature on different types of scoring models, the aspects of essay quality studied in prior research, and the implementation and evaluation of a scoring engine. Part 4 offers a broader view of the field inclusive of some neighboring areas, and Part \ref{part5} closes with summary and discussion. This book grew out of a week-long course on automated evaluation of language production at the North American Summer School for Logic, Language, and Information (NASSLLI), attended by advanced undergraduates and early-stage graduate students from a variety of disciplines. Teachers of natural language processing, in particular, will find that the book offers a useful foundation for a supplemental module on automated scoring. Professionals and students in linguistics, applied linguistics, educational technology, and other related disciplines will also find the material here useful.

Empirical methods are means to answering methodological questions of empirical sciences by statistical techniques. The methodological questions addressed in this book include the problems of validity, reliability, and significance. In the case of machine learning, these correspond to the questions of whether a model predicts what it purports to predict, whether a model's performance is consistent across replications, and whether a performance difference between two models is due to chance, respectively. The goal of this book is to answer these questions by concrete statistical tests that can be applied to assess validity, reliability, and significance of data annotation and machine learning prediction in the fields of NLP and data science. Our focus is on model-based empirical methods where data annotations and model predictions are treated as training data for interpretable probabilistic models from the well-understood families of generalized additive models (GAMs) and linear mixed effects models (LMEMs). Based on the interpretable parameters of the trained GAMs or LMEMs, the book presents model-based statistical tests such as a validity test that allows detecting circular features that circumvent learning. Furthermore, the book discusses a reliability coefficient using variance decomposition based on random effect parameters of LMEMs. Last, a significance test based on the likelihood ratio of nested LMEMs trained on the performance scores of two machine learning models is shown to naturally allow the inclusion of variations in meta-parameter settings into hypothesis testing, and further facilitates a refined system comparison conditional on properties of input data. This book can be used as an introduction to empirical methods for machine learning in general, with a special focus on applications in NLP and data science. The book is self-contained, with an appendix on the mathematical background on GAMs and LMEMs, and with an accompanying webpage including R code to replicate experiments presented in the book.

Learning to rank refers to machine learning techniques for training a model in a ranking task. Learning to rank is useful for many applications in information retrieval, natural language processing, and data mining. Intensive studies have been conducted on its problems recently, and significant progress has been made. This lecture gives an introduction to the area including the fundamental problems, major approaches, theories, applications, and future work. The author begins by showing that various ranking problems in information retrieval and natural language processing can be formalized as two basic ranking tasks, namely ranking creation (or simply ranking) and ranking aggregation. In ranking creation, given a request, one wants to generate a ranking list of offerings based on the features derived from the request and the offerings. In ranking aggregation, given a request, as well as a number of ranking lists of offerings, one wants to generate a new ranking list of the offerings. Ranking creation (or ranking) is the major problem in learning to rank. It is usually formalized as a supervised learning task. The author gives detailed explanations on learning for ranking creation and ranking aggregation, including training and testing, evaluation, feature creation, and major approaches. Many methods have been proposed for ranking creation. The methods can be categorized as the pointwise, pairwise, and listwise approaches according to the loss functions they employ. They can also be categorized according to the techniques they employ, such as the SVM based, Boosting based, and Neural Network based approaches. The author also introduces some popular learning to rank methods in details. These include: PRank, OC SVM, McRank, Ranking SVM, IR SVM, GBRank, RankNet, ListNet & ListMLE, AdaRank, SVM MAP, SoftRank, LambdaRank, LambdaMART, Borda Count, Markov Chain, and CRanking. The author explains several example applications of learning to rank including web search, collaborative filtering, definition search, keyphrase extraction, query dependent summarization, and re-ranking in machine translation. A formulation of learning for ranking creation is given in the statistical learning framework. Ongoing and future research directions for learning to rank are also discussed. Table of Contents: Learning to Rank / Learning for Ranking Creation / Learning for Ranking Aggregation / Methods of Learning to Rank / Applications of Learning to Rank / Theory of Learning to Rank / Ongoing and Future Work