Hang Li

Matching, which is to measure the relevance of a document to a query or interest of a user to an item, is a key problem in both search and recommendation. Machine learning has been exploited to address the problem and efforts have been made to develop deep learning techniques for matching tasks in search and recommendation. With the availability of a large amount of data, powerful computational resources, and advanced deep learning techniques, deep learning for matching now becomes the state-of-the-art technology for search and recommendation. The key to the success of the deep learning approach is its strong ability in learning of representations and generalization of matching patterns from data. This survey gives a systematic and comprehensive introduction to the deep matching models for search and recommendation. First, it gives a unified view of matching in search and recommendation and the solutions from the two fields can be compared in one framework. Then, the survey categorizes the current deep learning solutions into two types: methods of representation learning and methods of matching function learning. The fundamental problems as well as the state-of-the-art solutions of query-document matching in search and user-item matching in recommendation are described. Deep Learning for Matching in Search and Recommendation aims to help researchers from both search and recommendation communities to get an in-depth understanding and insight into the spaces, stimulate more ideas and discussions, and promote developments of new technologies. As matching is not limited to search and recommendation, the technologies introduced here can be generalized into a more general task of matching between objects from two spaces.

This book provides a comprehensive and systematic introduction to the principal machine learning methods, covering both supervised and unsupervised learning methods. It discusses essential methods of classification and regression in supervised learning, such as decision trees, perceptrons, support vector machines, maximum entropy models, logistic regression models and multiclass classification, as well as methods applied in supervised learning, like the hidden Markov model and conditional random fields. In the context of unsupervised learning, it examines clustering and other problems as well as methods such as singular value decomposition, principal component analysis and latent semantic analysis. As a fundamental book on machine learning, it addresses the needs of researchers and students who apply machine learning as an important tool in their research, especially those in fields such as information retrieval, natural language processing and text data mining. In order to understand the concepts and methods discussed, readers are expected to have an elementary knowledge of advanced mathematics, linear algebra and probability statistics. The detailed explanations of basic principles, underlying concepts and algorithms enable readers to grasp basic techniques, while the rigorous mathematical derivations and specific examples included offer valuable insights into machine learning.

This book provides a comprehensive and systematic introduction to the principal machine learning methods, covering both supervised and unsupervised learning methods. It discusses essential methods of classification and regression in supervised learning, such as decision trees, perceptrons, support vector machines, maximum entropy models, logistic regression models and multiclass classification, as well as methods applied in supervised learning, like the hidden Markov model and conditional random fields. In the context of unsupervised learning, it examines clustering and other problems as well as methods such as singular value decomposition, principal component analysis and latent semantic analysis. As a fundamental book on machine learning, it addresses the needs of researchers and students who apply machine learning as an important tool in their research, especially those in fields such as information retrieval, natural language processing and text data mining. In order to understand the concepts and methods discussed, readers are expected to have an elementary knowledge of advanced mathematics, linear algebra and probability statistics. The detailed explanations of basic principles, underlying concepts and algorithms enable readers to grasp basic techniques, while the rigorous mathematical derivations and specific examples included offer valuable insights into machine learning.

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

Semantic Matching in Search is a systematic and detailed introduction to newly developed machine learning technologies for query document matching (semantic matching) in search, particularly in web search. It focuses on the fundamental problems, as well as the state-of-the-art solutions of query document matching on form aspect, phrase aspect, word sense aspect, topic aspect, and structure aspect. Matching between query and document is not limited to search, and similar problems can be found in question answering, online advertising, cross-language information retrieval, machine translation, recommender systems, link prediction, image annotation, drug design, and other applications where one is faced with the general task of matching between objects from two different spaces. The technologies introduced in this monograph can be generalized into more general machine learning techniques, which are referred to as learning to match in this survey. It is hoped that the ideas and solutions explained in Semantic Matching in Search may motivate industrial practitioners to turn the research results into products. The methods introduced and the discussions around them should also stimulate academic researchers to find new research directions and approaches.