Since the seminal work of Mikolov et al., word embeddings have become the preferred word representations for many natural language processing tasks. Document similarity measures extracted from word embeddings, such as the soft cosine measure (SCM) and the Word Mover's Distance (WMD), were reported to achieve state-of-the-art performance on semantic text similarity and text classification.
Despite the strong performance of the WMD on text classification and semantic text similarity, its super-cubic average time complexity is impractical. The SCM has quadratic worst-case time complexity, but its performance on text classification has never been compared with the WMD. Recently, two word embedding regularization techniques were shown to reduce storage and memory costs, and to improve training speed, document processing speed, and task performance on word analogy, word similarity, and semantic text similarity. However, the effect of these techniques on text classification has not yet been studied.
In our work, we investigate the individual and joint effect of the two word embedding regularization techniques on the document processing speed and the task performance of the SCM and the WMD on text classification. For evaluation, we use the $k$NN classifier and six standard datasets: BBCSPORT, TWITTER, OHSUMED, REUTERS-21578, AMAZON, and 20NEWS.
We show 39% average $k$NN test error reduction with regularized word embeddings compared to non-regularized word embeddings. We describe a practical procedure for deriving such regularized embeddings through Cholesky factorization. We also show that the SCM with regularized word embeddings significantly outperforms the WMD on text classification and is over 10,000 times faster.
In many applications of natural language processing (NLP) it is necessary to determine the likelihood of a given word combination. For example, a speech recognizer may need to determine which of the two word combinations ``eat a peach'' and ``eat a beach'' is more likely. Statistical NLP methods determine the likelihood of a word combination from its frequency in a training corpus. However, the nature of language is such that many word combinations are infrequent and do not occur in any given corpus. In this work we propose a method for estimating the probability of such previously unseen word…
Huazhong University of Science and Technology（华中科技大学）；Cornell University（康奈尔大学）
Separable Non-negative Matrix Factorization (SNMF) is an important method for topic modeling, where "separable" assumes every topic contains at least one anchor word, defined as a word that has non-zero probability only on that topic. SNMF focuses on the word co-occurrence patterns to reveal topics by two steps: anchor word selection and topic recovery. The quality of the anchor words strongly influences the quality of the extracted topics. Existing anchor word selection algorithm is to greedily find an approximate convex hull in a high-dimensional word co-occurrence space. In this work, we pr…
Japan Advanced Institute of Science and Technology (JAIST)（日本先进科学技术学院（JAIST））；Toshiba Research & Development Center, Japan（日本东芝研发中心）
Different word embedding models capture different aspects of linguistic properties. This inspired us to propose a model (M-MaxLSTM-CNN) for employing multiple sets of word embeddings for evaluating sentence similarity/relation. Representing each word by multiple word embeddings, the MaxLSTM-CNN encoder generates a novel sentence embedding. We then learn the similarity/relation between our sentence embeddings via Multi-level comparison. Our method M-MaxLSTM-CNN consistently shows strong performances in several tasks (i.e., measure textual similarity, identify paraphrase, recognize textual entai…
University of Southern California（南加州大学）；National University of Singapore（新加坡国立大学）
Word Mover's Distance (WMD) computes the distance between words and models text similarity with the moving cost between words in two text sequences. Yet, it does not offer good performance in sentence similarity evaluation since it does not incorporate word importance and fails to take inherent contextual and structural information in a sentence into account. An improved WMD method using the syntactic parse tree, called Syntax-aware Word Mover's Distance (SynWMD), is proposed to address these two shortcomings in this work. First, a weighted graph is built upon the word co-occurrence statistics…
Similarity measures based purely on word embeddings are comfortably competing with much more sophisticated deep learning and expert-engineered systems on unsupervised semantic textual similarity (STS) tasks. In contrast to commonly used geometric approaches, we treat a single word embedding as e.g. 300 observations from a scalar random variable. Using this paradigm, we first illustrate that similarities derived from elementary pooling operations and classic correlation coefficients yield excellent results on standard STS benchmarks, outperforming many recently proposed methods while being much…
We compare four similarity-based estimation methods against back-off and maximum-likelihood estimation methods on a pseudo-word sense disambiguation task in which we controlled for both unigram and bigram frequency. The similarity-based methods perform up to 40% better on this particular task. We also conclude that events that occur only once in the training set have major impact on similarity-based estimates.
This dissertation explores the linguistic and computational aspects of the meaning relations that can hold between two or more complex linguistic expressions (phrases, clauses, sentences, paragraphs). In particular, it focuses on Paraphrasing, Textual Entailment, Contradiction, and Semantic Similarity. In Part I: "Similarity at the Level of Words and Phrases", I study the Distributional Hypothesis (DH) and explore several different methodologies for quantifying semantic similarity at the levels of words and short phrases. In Part II: "Paraphrase Typology and Paraphrase Identification", I focus…
Semantic textual similarity is one of the open research challenges in the field of Natural Language Processing. Extensive research has been carried out in this field and near-perfect results are achieved by recent transformer-based models in existing benchmark datasets like the STS dataset and the SICK dataset. In this paper, we study the sentences in these datasets and analyze the sensitivity of various word embeddings with respect to the complexity of the sentences. We build a complex sentences dataset comprising of 50 sentence pairs with associated semantic similarity values provided by 15 …
We describe a method for automatic word sense disambiguation using a text corpus and a machine-readable dictionary (MRD). The method is based on word similarity and context similarity measures. Words are considered similar if they appear in similar contexts; contexts are similar if they contain similar words. The circularity of this definition is resolved by an iterative, converging process, in which the system learns from the corpus a set of typical usages for each of the senses of the polysemous word listed in the MRD. A new instance of a polysemous word is assigned the sense associated with…
Recent literature suggests that averaged word vectors followed by simple post-processing outperform many deep learning methods on semantic textual similarity tasks. Furthermore, when averaged word vectors are trained supervised on large corpora of paraphrases, they achieve state-of-the-art results on standard STS benchmarks. Inspired by these insights, we push the limits of word embeddings even further. We propose a novel fuzzy bag-of-words (FBoW) representation for text that contains all the words in the vocabulary simultaneously but with different degrees of membership, which are derived fro…
Various applications in the areas of computational linguistics and artificial intelligence employ semantic similarity to solve challenging tasks, such as word sense disambiguation, text classification, information retrieval, machine translation, and document clustering. Previous work on semantic similarity followed a mono-relational approach using mostly the taxonomic relation "ISA". This paper explores the benefits of using all types of non-taxonomic relations in large linked data, such as WordNet knowledge graph, to enhance existing semantic similarity and relatedness measures. We propose a …
In natural language processing, word-sense disambiguation (WSD) is an open problem concerned with identifying the correct sense of words in a particular context. To address this problem, we introduce a novel knowledge-based WSD system. We suggest the adoption of two methods in our system. First, we suggest a novel method to encode the word vector representation by considering the graphical semantic relationships from the lexical knowledge-base. Second, we propose a method for extracting the contextual words from the text for analyzing an ambiguous word based on the similarity of word vector re…
University of Oslo（奥斯陆大学）；University of Hamburg（汉堡大学）；Skolkovo Institute of Science and Technology（斯科尔科沃科学技术学院）
We present path2vec, a new approach for learning graph embeddings that relies on structural measures of pairwise node similarities. The model learns representations for nodes in a dense space that approximate a given user-defined graph distance measure, such as e.g. the shortest path distance or distance measures that take information beyond the graph structure into account. Evaluation of the proposed model on semantic similarity and word sense disambiguation tasks, using various WordNet-based similarity measures, show that our approach yields competitive results, outperforming strong graph em…
A large body of research into semantic textual similarity has focused on constructing state-of-the-art embeddings using sophisticated modelling, careful choice of learning signals and many clever tricks. By contrast, little attention has been devoted to similarity measures between these embeddings, with cosine similarity being used unquestionably in the majority of cases. In this work, we illustrate that for all common word vectors, cosine similarity is essentially equivalent to the Pearson correlation coefficient, which provides some justification for its use. We thoroughly characterise cases…
Word vector representations open up new opportunities to extract useful information from unstructured text. Defining a word as a vector made it easy for the machine learning algorithms to understand a text and extract information from. Word vector representations have been used in many applications such word synonyms, word analogy, syntactic parsing, and many others. GloVe, based on word contexts and matrix vectorization, is an ef-fective vector-learning algorithm. It improves on previous vector-learning algorithms. However, the GloVe model fails to explicitly consider the order in which words…
Resolution of lexical ambiguity, commonly termed ``word sense disambiguation'', is expected to improve the analytical accuracy for tasks which are sensitive to lexical semantics. Such tasks include machine translation, information retrieval, parsing, natural language understanding and lexicography. Reflecting the growth in utilization of machine readable texts, word sense disambiguation techniques have been explored variously in the context of corpus-based approaches. Within one corpus-based framework, that is the similarity-based method, systems use a database, in which example sentences are …
Retrofitting techniques, which inject external resources into word representations, have compensated the weakness of distributed representations in semantic and relational knowledge between words. Implicitly retrofitting word vectors by expansional technique outperforms retrofitting in word similarity tasks with word vector generalization. In this paper, we propose unsupervised extrofitting: expansional retrofitting (extrofitting) without external semantic lexicons. We also propose deep extrofitting: in-depth stacking of extrofitting and further combinations of extrofitting with retrofitting. …
University of Illinois at Urbana Champaign（伊利诺伊大学厄巴纳-香槟分校）
Real-valued word representations have transformed NLP applications; popular examples are word2vec and GloVe, recognized for their ability to capture linguistic regularities. In this paper, we demonstrate a {\em very simple}, and yet counter-intuitive, postprocessing technique -- eliminate the common mean vector and a few top dominating directions from the word vectors -- that renders off-the-shelf representations {\em even stronger}. The postprocessing is empirically validated on a variety of lexical-level intrinsic tasks (word similarity, concept categorization, word analogy) and sentence-lev…
In this paper, we introduce a new WordNet based similarity metric, SenSim, which incorporates sentiment content (i.e., degree of positive or negative sentiment) of the words being compared to measure the similarity between them. The proposed metric is based on the hypothesis that knowing the sentiment is beneficial in measuring the similarity. To verify this hypothesis, we measure and compare the annotator agreement for 2 annotation strategies: 1) sentiment information of a pair of words is considered while annotating and 2) sentiment information of a pair of words is not considered while anno…
There are at least two kinds of similarity. Relational similarity is correspondence between relations, in contrast with attributional similarity, which is correspondence between attributes. When two words have a high degree of attributional similarity, we call them synonyms. When two pairs of words have a high degree of relational similarity, we say that their relations are analogous. For example, the word pair mason:stone is analogous to the pair carpenter:wood. This paper introduces Latent Relational Analysis (LRA), a method for measuring relational similarity. LRA has potential applications…
This paper introduces Latent Relational Analysis (LRA), a method for measuring relational similarity. LRA has potential applications in many areas, including information extraction, word sense disambiguation, machine translation, and information retrieval. Relational similarity is correspondence between relations, in contrast with attributional similarity, which is correspondence between attributes. When two words have a high degree of attributional similarity, we call them synonyms. When two pairs of words have a high degree of relational similarity, we say that their relations are analogous.…
The inclusion of semantic information in any similarity measures improves the efficiency of the similarity measure and provides human interpretable results for further analysis. The similarity calculation method that focuses on features related to the text's words only, will give less accurate results. This paper presents three different methods that not only focus on the text's words but also incorporates semantic information of texts in their feature vector and computes semantic similarities. These methods are based on corpus-based and knowledge-based methods, which are: cosine similarity us…
Sentence similarity is considered the basis of many natural language tasks such as information retrieval, question answering and text summarization. The semantic meaning between compared text fragments is based on the words semantic features and their relationships. This article reviews a set of word and sentence similarity measures and compares them on benchmark datasets. On the studied datasets, results showed that hybrid semantic measures perform better than both knowledge and corpus based measures.
There have been several efforts to extend distributional semantics beyond individual words, to measure the similarity of word pairs, phrases, and sentences (briefly, tuples; ordered sets of words, contiguous or noncontiguous). One way to extend beyond words is to compare two tuples using a function that combines pairwise similarities between the component words in the tuples. A strength of this approach is that it works with both relational similarity (analogy) and compositional similarity (paraphrase). However, past work required hand-coding the combination function for different tasks. The m…
In this paper, we discuss a method for identifying a seed word that would best represent a class of named entities in a graphical representation of words and their similarities. Word networks, or word graphs, are representations of vectorized text where nodes are the words encountered in a corpus, and the weighted edges incident on the nodes represent how similar the words are to each other. We intend to build a bilingual word graph and identify seed words through community analysis that would be best used to segment a graph according to its named entities, therefore providing an unsupervised …
Word embedding, specially with its recent developments, promises a quantification of the similarity between terms. However, it is not clear to which extent this similarity value can be genuinely meaningful and useful for subsequent tasks. We explore how the similarity score obtained from the models is really indicative of term relatedness. We first observe and quantify the uncertainty factor of the word embedding models regarding to the similarity value. Based on this factor, we introduce a general threshold on various dimensions which effectively filters the highly related terms. Our evaluati…
University Sidi Mohamed Ben Abdellah (USMBA)（西迪·穆罕默德·本·阿卜杜拉大学）；Faculty of Science Dhar EL Mahraz (FSDM)（达尔·马赫拉兹科学学院）
Representation of semantic information contained in the words is needed for any Arabic Text Mining applications. More precisely, the purpose is to better take into account the semantic dependencies between words expressed by the co-occurrence frequencies of these words. There have been many proposals to compute similarities between words based on their distributions in contexts. In this paper, we compare and contrast the effect of two preprocessing techniques applied to Arabic corpus: Rootbased (Stemming), and Stem-based (Light Stemming) approaches for measuring the similarity between Arabic w…
University of Illinois College of Medicine（伊利诺伊大学医学院）
Neural embeddings are a popular set of methods for representing words, phrases or text as a low dimensional vector (typically 50-500 dimensions). However, it is difficult to interpret these dimensions in a meaningful manner, and creating neural embeddings requires extensive training and tuning of multiple parameters and hyperparameters. We present here a simple unsupervised method for representing words, phrases or text as a low dimensional vector, in which the meaning and relative importance of dimensions is transparent to inspection. We have created a near-comprehensive vector representation…
There is a great deal of work in cognitive psychology, linguistics, and computer science, about using word (or phrase) frequencies in context in text corpora to develop measures for word similarity or word association, going back to at least the 1960s. The goal of this chapter is to introduce the normalizedis a general way to tap the amorphous low-grade knowledge available for free on the Internet, typed in by local users aiming at personal gratification of diverse objectives, and yet globally achieving what is effectively the largest semantic electronic database in the world. Moreover, this d…
Levering data on social media, such as Twitter and Facebook, requires information retrieval algorithms to become able to relate very short text fragments to each other. Traditional text similarity methods such as tf-idf cosine-similarity, based on word overlap, mostly fail to produce good results in this case, since word overlap is little or non-existent. Recently, distributed word representations, or word embeddings, have been shown to successfully allow words to match on the semantic level. In order to pair short text fragments - as a concatenation of separate words - an adequate distributed…
ICMC-Universidade de São Paulo (USP)（圣保罗大学计算与数学科学中心（ICMC））；DC-Universidade Federal de São Carlos (UFSCar)（圣卡洛斯联邦大学计算系（DC））；Universidade Federal de Mato Grosso do Sul (UFMS)（南马托格罗索联邦大学）；Ericsson AB / Blekinge Institute of Technology（爱立信公司/布莱金厄理工学院）；INF-Universidade Federal de Goiás (UFG)（戈亚斯联邦大学信息系（INF））；Tribunal de Justiça do Estado de Goiás（戈亚斯州法院）
Bug localization (BL) from the bug report is the strategic activity of the software maintaining process. Because BL is a costly and tedious activity, BL techniques information retrieval-based and machine learning-based could aid software engineers. We propose a method for BUg Localization with word embeddings and Network Regularization (BULNER). The preliminary results suggest that BULNER has better performance than two state-of-the-art methods.
Research Training Group AIPHES（AIPHES研究培训组）；Institute for Computational Linguistics（计算语言学研究所）；Heidelberg University（海德堡大学）；Qatar Computing Research Institute, HBKU（哈马德·本·哈利法大学卡塔尔计算研究所）
We describe our system for finding good answers in a community forum, as defined in SemEval-2016, Task 3 on Community Question Answering. Our approach relies on several semantic similarity features based on fine-tuned word embeddings and topics similarities. In the main Subtask C, our primary submission was ranked third, with a MAP of 51.68 and accuracy of 69.94. In Subtask A, our primary submission was also third, with MAP of 77.58 and accuracy of 73.39.
Eliciting semantic similarity between concepts in the biomedical domain remains a challenging task. Recent approaches founded on embedding vectors have gained in popularity as they risen to efficiently capture semantic relationships The underlying idea is that two words that have close meaning gather similar contexts. In this study, we propose a new neural network model named MeSH-gram which relies on a straighforward approach that extends the skip-gram neural network model by considering MeSH (Medical Subject Headings) descriptors instead words. Trained on publicly available corpus PubMed MED…
To measure the similarity of two documents in the bag-of-words (BoW) vector representation, different term weighting schemes are used to improve the performance of cosine similarity---the most widely used inter-document similarity measure in text mining. In this paper, we identify the shortcomings of the underlying assumptions of term weighting in the inter-document similarity measurement task; and provide a more fit-to-the-purpose alternative. Based on this new assumption, we introduce a new simple but effective similarity measure which does not require explicit term weighting. The proposed m…
National Institute of Technology, Durgapur（杜尔加普尔国立理工学院）；Amdocs；Indian Statistical Institute（印度统计学院）
Information content (IC) based measures for finding semantic similarity is gaining preferences day by day. Semantics of concepts can be highly characterized by information theory. The conventional way for calculating IC is based on the probability of appearance of concepts in corpora. Due to data sparseness and corpora dependency issues of those conventional approaches, a new corpora independent intrinsic IC calculation measure has evolved. In this paper, we mainly focus on such intrinsic IC model and several topological aspects of the underlying ontology. Accuracy of intrinsic IC calculation …
Latent semantic representations of words or paragraphs, namely the embeddings, have been widely applied to information retrieval (IR). One of the common approaches of utilizing embeddings for IR is to estimate the document-to-query (D2Q) similarity in their embeddings. As words with similar syntactic usage are usually very close to each other in the embeddings space, although they are not semantically similar, the D2Q similarity approach may suffer from the problem of "multiple degrees of similarity". To this end, this paper proposes a novel approach that estimates a semantic relevance score (…
National Center for Biotechnology Information（国家生物技术信息中心）；National Library of Medicine, National Institutes of Health（美国国家医学图书馆，美国国立卫生研究院）
The main approach of traditional information retrieval (IR) is to examine how many words from a query appear in a document. A drawback of this approach, however, is that it may fail to detect relevant documents where no or only few words from a query are found. The semantic analysis methods such as LSA (latent semantic analysis) and LDA (latent Dirichlet allocation) have been proposed to address the issue, but their performance is not superior compared to common IR approaches. Here we present a query-document similarity measure motivated by the Word Mover's Distance. Unlike other similarity me…
The social media craze is on an ever increasing spree, and people are connected with each other like never before, but these vast connections are visually unexplored. We propose a methodology Twigraph to explore the connections between persons using their Twitter profiles. First, we propose a hybrid approach of recommending social media profiles, articles, and advertisements to a user.The profiles are recommended based on the similarity score between the user profile, and profile under evaluation. The similarity between a set of profiles is investigated by finding the top influential words thu…
Words and phrases acquire meaning from the way they are used in society, from their relative semantics to other words and phrases. For computers the equivalent of `society' is `database,' and the equivalent of `use' is `way to search the database.' We present a new theory of similarity between words and phrases based on information distance and Kolmogorov complexity. To fix thoughts we use the world-wide-web as database, and Google as search engine. The method is also applicable to other search engines and databases. This theory is then applied to construct a method to automatically extract si…
A major difficulty in applying word vector embeddings in IR is in devising an effective and efficient strategy for obtaining representations of compound units of text, such as whole documents, (in comparison to the atomic words), for the purpose of indexing and scoring documents. Instead of striving for a suitable method for obtaining a single vector representation of a large document of text, we rather aim for developing a similarity metric that makes use of the similarities between the individual embedded word vectors in a document and a query. More specifically, we represent a document and …
In this report, we experimented with several concepts regarding text streams analysis. We tested an implementation of Incremental Sparse TF-IDF (IS-TFIDF) and Incremental Cosine Similarity (ICS) with the use of bipartite graphs. We are using bipartite graphs - one type of node are documents, and the other type of nodes are words - to know what documents are affected with a word arrival at the stream (the neighbors of the word in the graph). Thus, with this information, we leverage optimized algorithms used for graph-based applications. The concept is similar to, for example, the use of hash ta…
We engineer an algorithm to solve the approximate dictionary matching problem. Given a list of words $\mathcal{W}$, maximum distance $d$ fixed at preprocessing time and a query word $q$, we would like to retrieve all words from $\mathcal{W}$ that can be transformed into $q$ with $d$ or less edit operations. We present data structures that support fault tolerant queries by generating an index. On top of that, we present a generalization of the method that eases memory consumption and preprocessing time significantly. At the same time, running times of queries are virtually unaffected. We are ab…
Recently, there is a surge of social recommendation, which leverages social relations among users to improve recommendation performance. However, in many applications, social relations are absent or very sparse. Meanwhile, the attribute information of users or items may be rich. It is a big challenge to exploit these attribute information for the improvement of recommendation performance. In this paper, we organize objects and relations in recommendation system as a heterogeneous information network, and introduce meta path based similarity measure to evaluate the similarity of users or items.…
