acl acl2013 acl2013-304 knowledge-graph by maker-knowledge-mining
Source: pdf
Author: Vasile Rus ; Mihai Lintean ; Rajendra Banjade ; Nobal Niraula ; Dan Stefanescu
Abstract: We present in this paper SEMILAR, the SEMantic simILARity toolkit. SEMILAR implements a number of algorithms for assessing the semantic similarity between two texts. It is available as a Java library and as a Java standalone ap-plication offering GUI-based access to the implemented semantic similarity methods. Furthermore, it offers facilities for manual se-mantic similarity annotation by experts through its component SEMILAT (a SEMantic simILarity Annotation Tool). 1
Reference: text
sentIndex sentText sentNum sentScore
1 SEMILAR implements a number of algorithms for assessing the semantic similarity between two texts. [sent-3, score-0.521]
2 It is available as a Java library and as a Java standalone ap-plication offering GUI-based access to the implemented semantic similarity methods. [sent-4, score-0.499]
3 Furthermore, it offers facilities for manual se-mantic similarity annotation by experts through its component SEMILAT (a SEMantic simILarity Annotation Tool). [sent-5, score-0.518]
4 org) includes implementations of a number of algorithms pro- posed over the last decade or so to address various instances of the general problem of text-to-text semantic similarity. [sent-9, score-0.152]
5 Semantic similarity is an approach to language understanding that is widely used in real applications. [sent-10, score-0.333]
6 Given such two texts, the paraphrase identification task is about automatically assessing whether Text A is a paraphrase of, i. [sent-14, score-0.522]
7 The example above is a positive instance, meaning that Text A is a paraphrase of Text B and vice versa. [sent-17, score-0.199]
8 The importance of semantic similarity in Natural Language Processing (NLP) is highlighted by the diversity of datasets and shared task evaluation campaigns (STECs) that have been proposed over the last decade (Dolan, Quirk, and Brockett, 2004; McCarthy & McNamara, 2008; Agirre et al. [sent-18, score-0.485]
9 Indeed, there is a need to identify and quantify semantic relations between texts in many applications. [sent-21, score-0.255]
10 For instance, paraphrase identification, an instance of the semantic similarity problem, is an important step in a number of applications including Natural Language Generation, Question Answering, and dialogue-based Intelligent Tutoring Systems. [sent-22, score-0.641]
11 In Natural Language Generation, paraphrases are a method to increase diversity of generated text (Iordanskaja et al. [sent-23, score-0.094]
12 In Question Answering, multiple answers that are paraphrases of each other could be considered as evidence for the correctness of the answer (Ibrahim et al. [sent-25, score-0.047]
13 , 2010; Lintean, 2011), paraphrase identification is useful to assess whether students’ articulated answers to deep questions (e. [sent-29, score-0.289]
14 Generally, the problem of semantic similarity between two texts, denoted text A and text B, is defined as quantifying and identifying the presence of semantic relations between the two texts, e. [sent-32, score-0.674]
15 to what extent text A has the same meaning as or is a paraphrase of text B (paraphrase relation; Dolan, Quirk, and Brockett, 2004). [sent-34, score-0.293]
16 Other semantic relations that have been investigated systematically in the recent past are entailment, i. [sent-35, score-0.138]
17 to what extent text A entails or logically infers text B (Dagan, Glickman, & Magnini, 2004), and elaboration, i. [sent-37, score-0.168]
18 c A2s0s1o3ci Aatsiosonc fioartio Cno fmorpu Ctoamtiopnuatalt Lioin gauli Lsitnicgsu,i psatgices 163–168, Semantic similarity can be broadly construed between texts of any size. [sent-43, score-0.45]
19 Depending on the granularity of the texts, we can talk about the following fundamental text-to-text similarity problems: word-to-word similarity, phrase-tophrase similarity, sentence-to-sentence similarity, paragraph-to-paragraph similarity, or document-to-document similarity. [sent-44, score-0.333]
20 Mixed combinations are also possible such as assessing the similarity of a word to a sentence or a sentence to a paragraph. [sent-45, score-0.412]
21 2 Motivation The problem of word-to-word similarity has been extensively studied over the past decades and a word-to-word similarity library (WordNet Similarity) has been developed by Pedersen and colleagues (Pedersen, Patwardhan, & Michelizzi, 2004). [sent-47, score-0.782]
22 Methods to assess the semantic similarity of larger texts, in particular sentences, have been proposed over the last decade (Corley and Mihalcea, 2005; Fernando & Stevenson, 2008; Rus, Lintean, Graesser, & McNamara 2009). [sent-48, score-0.53]
23 Androutsopoulos & Malakasiotis (2010) compiled a survey of methods for paraphrasing and entailment semantic relation identification at sentence level. [sent-49, score-0.272]
24 Despite all the proposed methods to assess semantic similarity between two texts, no semantic similarity library or toolkit, similar to the WordNet library for word-to-word similarity, exists for larger texts. [sent-50, score-1.043]
25 Given the importance of semantic similarity, there is an acute need for such a library and toolkit. [sent-51, score-0.166]
26 The developed SEMILAR library and toolkit presented here fulfill this need. [sent-52, score-0.122]
27 In particular, the development of the semantic similarity toolkit SEMILAR has been motivated by the need for an integrated environment that would provide: easy access to implementations of various semantic similarity approaches from the same user-friendly interface and/or library. [sent-53, score-0.949]
28 easy access to semantic similarity methods that work at different levels of text granularity: word-to-word, sentence-to-sentence, paragraph-to-paragraph, document-todocument, or a combination (SEMILAR in- tegrates word-to-word similarity measures). [sent-54, score-0.822]
29 a common environment for that allows systematic and fair comparison of semantic similarity methods. [sent-56, score-0.442]
30 facilities to manually annotate texts with semantic similarity relations using a graphical user interface that make such annotations easier for experts (this component is called SEMILAT component - a SEMantic similarity Annotation Tool). [sent-57, score-1.152]
31 SEMILAR is thus a one-stop-shop for investigating, annotating, and authoring methods for the semantic similarity of texts of any level of granularity. [sent-58, score-0.603]
32 3 SEMILAR: Toolkit The Semantic Similarity The authors of the SEMILAR toolkit (see Figure 1) have been involved in assessing the semantic 164 similarity of texts for more than a decade. [sent-59, score-0.703]
33 During this time, they have conducted a careful requirements analysis for an integrated software toolkit that would integrate various methods for semantic similarity assessment. [sent-60, score-0.507]
34 We briefly present the components of SEMILAR next and then describe in more detail the core component of SEMILAR, i. [sent-62, score-0.046]
35 the set of semantic similarity methods that are currently available. [sent-64, score-0.442]
36 It should be noted that we are continuously adding new semantic similarity methods and features to SEMILAR. [sent-65, score-0.469]
37 The SEMILAR toolkit includes the following components: project management; data viewbrowsing-visualization; preprocessing (e. [sent-66, score-0.065]
38 the major text-to-text similarity algorithms currently available in SEMILAR. [sent-73, score-0.333]
39 4 The Semantic Similarity Available in SEMILAR Methods The core component of SEMILAR is a set of text-to-text semantic similarity methods. [sent-74, score-0.488]
40 We have implemented methods that handle both unidirectional similarity measures as well as bidirectional similarity measures. [sent-75, score-0.815]
41 For instance, the semantic relation of entailment between two texts is unidirectional (a text T logically entails a hypothesis text H but H does not entail T) while the paraphrase relation is bidirectional (text A has same meaning as text B and vice versa). [sent-76, score-0.808]
42 Given two texts, the simplest method to assess their semantic similarity is to compute lexical overlap, i. [sent-78, score-0.517]
43 Indeed, a closer look at lexical overlap reveals a number of parameters that turns the simple lexical overlap problem into a large space of possibilities. [sent-82, score-0.276]
44 A total of 3,456 variants of lexical overlap can be generated by different parameter settings in SEMILAR. [sent-89, score-0.138]
45 Lintean (201 1) has shown that performance on lexical overlap methods on the tasks of paraphrase identification and textual entailment tasks can vary significantly depending on the selected parameters. [sent-90, score-0.496]
46 Some lexical overlap variations lead to performance results rivaling more sophisticated, state-of-the-art methods. [sent-91, score-0.138]
47 It should be noted that the overlap category of methods can be extended to include N-gram overlap methods (see the N-gram overlap methods proposed by the Machine Translation community such as BLEU and METEOR). [sent-92, score-0.351]
48 SEMI- LAR offers bigram and unigram overlap methods including the BLEU and METEOR scores. [sent-93, score-0.108]
49 A natural approach to text-to-text similarity methods is to rely on word-to-word similarity measures. [sent-94, score-0.666]
50 Many of the methods presented next compute the similarity of larger texts using individual word similarities. [sent-95, score-0.45]
51 Mihalcea, Corley, & Strappavara (2006; MCS) proposed a greedy method based on wordto-word similarity measures. [sent-96, score-0.333]
52 For each word in text A (or B) the maximum similarity score to any word in the other text B (or A) is used. [sent-97, score-0.427]
53 sim(T 1,T2)1(w{T1}max{Sim(w,T2)*idf(w)} 2 w{T1}idf(w) w{T2}max{Sim(w,T1)*idf(w)} ) w{T2}idf(w) The word-to-word similarity function sim(w, T) in the equation above can be instantiated to any word-to-word similarity measure (e. [sent-99, score-0.696]
54 The vast majority of word-to-word similarity measures that rely on WordNet are concept-toconcept measures and to be able to use them one must map words in the input texts onto concepts in WordNet, i. [sent-102, score-0.576]
55 We label the former method as ONE (sense one), whereas the latter is labeled as ALL-MAX or ALL-AVG (all senses maximum score or all senses average score, respectively). [sent-106, score-0.06]
56 Furthermore, most WordNet-based measures only work within a part-of-speech category, e. [sent-107, score-0.063]
57 Rus and Lintean (2012; Rus-LinteanOptimal Matching or ROM) proposed an optimal solution for text-to-text similarity based on word-to-word similarity measures. [sent-111, score-0.697]
58 The optimal lexical matching is based on the optimal assignment problem, a fundamental combinatorial optimization problem which consists of finding a maximum weight matching in a weighted bipartite graph. [sent-112, score-0.221]
59 Given a weighted complete bipartite graph , where edge has weight , the optimal assignment problem is to find a matching M from X to Y with maximum weight. [sent-113, score-0.16]
60 words in text A in our case, to a set of jobs (words in text B in our case) based on the expertise level, measured by of each worker at each job. [sent-116, score-0.129]
61 By adding dummy workers or jobs we may assume that X and Y have the same size, n, and can be viewed as and Y = . [sent-117, score-0.067]
62 In the semantic similarity case, the weight is the word-to-word similarity between a word x , in text A and a word y in text B. [sent-118, score-0.869]
63 The assignment problem can also be stated as finding a permutation of { 1, 2, 3, , n} for which is maximum. [sent-119, score-0.072]
64 The Kuhn-Munkres algorithm (Kuhn, 1955) can find a solution to the optimum assignment problem in polynomial time. [sent-121, score-0.1]
65 Each text is regarded as a graph with words as nodes/vertices and syntactic dependencies as edges. [sent-124, score-0.047]
66 The subsumption score reflects how much a text is subsumed or contained by another. [sent-125, score-0.123]
67 The equation below provides the overall subsumption score, which can be averaged both ways to compute a similarity score, as opposed to just the subsumption score, between the two texts. [sent-126, score-0.515]
68 The match function in the equation can be any word-to-word similarity measure including simple word match, WordNet similarity measures, LSA, or LDA-based similarity measures. [sent-128, score-1.029]
69 Fernando and Stevenson (FST; 2008) proposed a method in which similarities among all pairs of words are taken into account for computing the similarity of two texts. [sent-129, score-0.368]
70 Each text is represented as a binary vector (1 – the word occurs in the text; 0 – the word does not occur in the text). [sent-130, score-0.047]
71 They use a similarity matrix operator W that contains word-to-word similarities between any two words. [sent-131, score-0.397]
72 , aWbT sim( a b ) | a || b | Each element wij represents the word-level semantic similarity between word ai in text A and word bj in text B. [sent-132, score-0.536]
73 Lintean and Rus (2010; weighted-LSA or wLSA) extensively studied methods for semantic similarity based on Latent Semantic Analysis (LSA; Landauer et al. [sent-134, score-0.468]
74 An LSA vector for larger texts can be derived by vector algebra, e. [sent-137, score-0.117]
75 The similarity of two texts A and B can be computed using the cosine (normalized dot product) of their LSA vectors. [sent-140, score-0.45]
76 All these versions of LSA-based text-to-text similarity measures are available in SEMILAR. [sent-143, score-0.396]
77 SEMILAR also includes a set of similarity measures based on the unsupervised method Latent Dirichlet Allocation (LDA; Blei, Ng, & Jordnan, 2003; Rus, Banjade, & Niraula, 2013). [sent-144, score-0.396]
78 LDA is a probabilistic generative model in which documents are viewed as distributions over a set of topics (θd - text d’s distribution over topics) and topics are distributions over words (φt – topic t’s distribution over words). [sent-145, score-0.311]
79 A first LDA-based semantic similarity measure among words would then be defined as a dotproduct between the corresponding vectors representing the contributions of each word to a topic (φt(w) – represents the probability of word w in topic t). [sent-147, score-0.442]
80 It should be noted that the contributions of each word to the topics does not constitute a distribution, i. [sent-148, score-0.097]
81 Assuming the number of topics T, then a simple word-to-word measure is defined by the formula below. [sent-151, score-0.07]
82 LDAw2w(w,v)Tt(w)t(v) t1 More global text-to-text similarity measures could be defined in several ways as detailed next. [sent-152, score-0.396]
83 Because in LDA a document is a distribution over topics, the similarity of two texts needs to be computed in terms of similarity of distributions. [sent-153, score-0.814]
84 KL(p,q)iT1pilogpqii If we replace p with θd (text/document d’s distribution over topics) and q with (text/document c’s distribution over topics) we obtain the KL distance between two documents (documents d and c in our example). [sent-155, score-0.062]
85 Also, the IR can be transformed into a symmetric similarity measure as in the following (Dagan, Lee, & Pereira, 1997): SIM(p,q) 10IR(c,d) The Hellinger and Manhattan distances between two distributions are two other options that avoid the shortcomings of the KL distance. [sent-160, score-0.409]
86 LDA similarity measures between two documents or texts c and d can also include similarity of topics. [sent-162, score-0.846]
87 That is, the text-to-text similarity is obtained multiplying the similarities between the distribution over topics (θd and θc) and distribution over words (φt1 and φt2). [sent-163, score-0.5]
88 The similarity of topics can be computed using the same methods illustrated above as the topics are distributions over words (for all the details see Rus, Banjade, & Niraula, 2013). [sent-164, score-0.504]
89 The last semantic similarity method presented in this paper is based on the Quadratic Assignment Problem (QAP). [sent-165, score-0.442]
90 The QAP method aims at finding an optimal assignment from words in text A to words in text B, based on individual wordto-word similarity measures, while simultaneously maximizing the match between the syntactic dependencies of the words. [sent-166, score-0.53]
91 n n n minQAP(F,D,B) i1j1fi,jd(i)(j) i1bi,(i) The fi,j term denotes the flow between facilities iand j which are placed at locations π(i) and π(j), respectively. [sent-170, score-0.202]
92 In our case, F and D describe dependencies between words in one sentence while B captures the word-to-word similarity between words in opposite sentences. [sent-172, score-0.333]
93 Also, we have weighted each term in the above formulation and instead of minimizing the sum we are maximizing it resulting in the formulation below. [sent-173, score-0.102]
94 For paraphrase identification, the QAP method provides best accuracy results (=77. [sent-175, score-0.199]
95 6%) on the test subset of the Microsoft Research Paraphrase corpus, one of the largest paraphrase datasets. [sent-176, score-0.199]
96 For a complete list of features, latest news, references, and updates of the SEMILAR toolkit along with downloadable resources including software and data files, the reader can visit this link: www. [sent-178, score-0.065]
97 Unsupervised construction of large paraphrase corpora: Exploiting massively parallel news sources. [sent-226, score-0.199]
98 A semantic similarity approach to paraphrase detection, Computational Linguistics UK (CLUK 2008) 11th Annual Research Colloquium. [sent-231, score-0.641]
99 Extracting structural paraphrases from aligned monolingual corpora In Proceedings of the Second International Workshop on Paraphrasing, (ACL 2003). [sent-247, score-0.047]
100 Lexical selection and paraphrase in a meaning-text generation model, Kluwer Academic. [sent-253, score-0.199]
wordName wordTfidf (topN-words)
[('semilar', 0.516), ('similarity', 0.333), ('rus', 0.304), ('lintean', 0.288), ('paraphrase', 0.199), ('mcnamara', 0.144), ('qap', 0.144), ('facilities', 0.139), ('niraula', 0.136), ('texts', 0.117), ('semantic', 0.109), ('banjade', 0.109), ('overlap', 0.108), ('kl', 0.096), ('entailment', 0.082), ('memphis', 0.082), ('assessing', 0.079), ('lsa', 0.077), ('subsumption', 0.076), ('assignment', 0.072), ('sim', 0.071), ('topics', 0.07), ('graesser', 0.067), ('toolkit', 0.065), ('measures', 0.063), ('brockett', 0.06), ('corley', 0.06), ('idf', 0.059), ('library', 0.057), ('unidirectional', 0.057), ('tutoring', 0.057), ('iordanskaja', 0.054), ('michelizzi', 0.054), ('semilat', 0.054), ('wordnet', 0.051), ('dagan', 0.051), ('lda', 0.049), ('dolan', 0.049), ('ibrahim', 0.048), ('mta', 0.048), ('stevenson', 0.048), ('paraphrases', 0.047), ('text', 0.047), ('magnini', 0.046), ('component', 0.046), ('assess', 0.045), ('identification', 0.045), ('options', 0.045), ('authoring', 0.044), ('malakasiotis', 0.044), ('latent', 0.043), ('allocation', 0.043), ('decade', 0.043), ('mccarthy', 0.043), ('pedersen', 0.042), ('androutsopoulos', 0.042), ('weighting', 0.041), ('elaboration', 0.04), ('logically', 0.04), ('locations', 0.037), ('formulation', 0.037), ('meteor', 0.037), ('moldovan', 0.037), ('funds', 0.037), ('kernels', 0.036), ('intelligent', 0.036), ('paraphrasing', 0.036), ('jobs', 0.035), ('similarities', 0.035), ('entails', 0.034), ('quirk', 0.034), ('eh', 0.033), ('colleagues', 0.033), ('mihalcea', 0.033), ('dirichlet', 0.033), ('ir', 0.033), ('glickman', 0.033), ('workers', 0.032), ('textual', 0.032), ('distribution', 0.031), ('distributions', 0.031), ('optimal', 0.031), ('senses', 0.03), ('equation', 0.03), ('patwardhan', 0.03), ('lexical', 0.03), ('matrix', 0.029), ('relations', 0.029), ('bipartite', 0.029), ('bidirectional', 0.029), ('weighted', 0.028), ('agirre', 0.028), ('collocation', 0.028), ('optimum', 0.028), ('fernando', 0.027), ('noted', 0.027), ('qualitative', 0.027), ('artificial', 0.026), ('flow', 0.026), ('extensively', 0.026)]
simIndex simValue paperId paperTitle
same-paper 1 1.0000001 304 acl-2013-SEMILAR: The Semantic Similarity Toolkit
Author: Vasile Rus ; Mihai Lintean ; Rajendra Banjade ; Nobal Niraula ; Dan Stefanescu
Abstract: We present in this paper SEMILAR, the SEMantic simILARity toolkit. SEMILAR implements a number of algorithms for assessing the semantic similarity between two texts. It is available as a Java library and as a Java standalone ap-plication offering GUI-based access to the implemented semantic similarity methods. Furthermore, it offers facilities for manual se-mantic similarity annotation by experts through its component SEMILAT (a SEMantic simILarity Annotation Tool). 1
2 0.19526277 43 acl-2013-Align, Disambiguate and Walk: A Unified Approach for Measuring Semantic Similarity
Author: Mohammad Taher Pilehvar ; David Jurgens ; Roberto Navigli
Abstract: Semantic similarity is an essential component of many Natural Language Processing applications. However, prior methods for computing semantic similarity often operate at different levels, e.g., single words or entire documents, which requires adapting the method for each data type. We present a unified approach to semantic similarity that operates at multiple levels, all the way from comparing word senses to comparing text documents. Our method leverages a common probabilistic representation over word senses in order to compare different types of linguistic data. This unified representation shows state-ofthe-art performance on three tasks: seman- tic textual similarity, word similarity, and word sense coarsening.
3 0.19096114 104 acl-2013-DKPro Similarity: An Open Source Framework for Text Similarity
Author: Daniel Bar ; Torsten Zesch ; Iryna Gurevych
Abstract: We present DKPro Similarity, an open source framework for text similarity. Our goal is to provide a comprehensive repository of text similarity measures which are implemented using standardized interfaces. DKPro Similarity comprises a wide variety of measures ranging from ones based on simple n-grams and common subsequences to high-dimensional vector comparisons and structural, stylistic, and phonetic measures. In order to promote the reproducibility of experimental results and to provide reliable, permanent experimental conditions for future studies, DKPro Similarity additionally comes with a set of full-featured experimental setups which can be run out-of-the-box and be used for future systems to built upon.
4 0.13239877 27 acl-2013-A Two Level Model for Context Sensitive Inference Rules
Author: Oren Melamud ; Jonathan Berant ; Ido Dagan ; Jacob Goldberger ; Idan Szpektor
Abstract: Automatic acquisition of inference rules for predicates has been commonly addressed by computing distributional similarity between vectors of argument words, operating at the word space level. A recent line of work, which addresses context sensitivity of rules, represented contexts in a latent topic space and computed similarity over topic vectors. We propose a novel two-level model, which computes similarities between word-level vectors that are biased by topic-level context representations. Evaluations on a naturallydistributed dataset show that our model significantly outperforms prior word-level and topic-level models. We also release a first context-sensitive inference rule set.
5 0.12101363 273 acl-2013-Paraphrasing Adaptation for Web Search Ranking
Author: Chenguang Wang ; Nan Duan ; Ming Zhou ; Ming Zhang
Abstract: Mismatch between queries and documents is a key issue for the web search task. In order to narrow down such mismatch, in this paper, we present an in-depth investigation on adapting a paraphrasing technique to web search from three aspects: a search-oriented paraphrasing model; an NDCG-based parameter optimization algorithm; an enhanced ranking model leveraging augmented features computed on paraphrases of original queries. Ex- periments performed on the large scale query-document data set show that, the search performance can be significantly improved, with +3.28% and +1.14% NDCG gains on dev and test sets respectively.
6 0.12093594 12 acl-2013-A New Set of Norms for Semantic Relatedness Measures
7 0.11477876 271 acl-2013-ParaQuery: Making Sense of Paraphrase Collections
8 0.11439791 291 acl-2013-Question Answering Using Enhanced Lexical Semantic Models
9 0.11134555 74 acl-2013-Building Comparable Corpora Based on Bilingual LDA Model
10 0.09737841 31 acl-2013-A corpus-based evaluation method for Distributional Semantic Models
11 0.09699145 297 acl-2013-Recognizing Partial Textual Entailment
12 0.091732778 222 acl-2013-Learning Semantic Textual Similarity with Structural Representations
13 0.087788589 262 acl-2013-Offspring from Reproduction Problems: What Replication Failure Teaches Us
14 0.085256256 93 acl-2013-Context Vector Disambiguation for Bilingual Lexicon Extraction from Comparable Corpora
15 0.084954813 96 acl-2013-Creating Similarity: Lateral Thinking for Vertical Similarity Judgments
16 0.08394482 376 acl-2013-Using Lexical Expansion to Learn Inference Rules from Sparse Data
17 0.082204908 109 acl-2013-Decipherment Complexity in 1:1 Substitution Ciphers
18 0.082052037 238 acl-2013-Measuring semantic content in distributional vectors
19 0.081603087 58 acl-2013-Automated Collocation Suggestion for Japanese Second Language Learners
20 0.079249501 139 acl-2013-Entity Linking for Tweets
topicId topicWeight
[(0, 0.189), (1, 0.086), (2, 0.061), (3, -0.177), (4, 0.014), (5, -0.092), (6, -0.066), (7, -0.036), (8, -0.055), (9, -0.035), (10, 0.041), (11, 0.109), (12, 0.031), (13, -0.044), (14, 0.128), (15, 0.022), (16, 0.006), (17, 0.006), (18, -0.042), (19, 0.04), (20, 0.028), (21, 0.012), (22, -0.045), (23, 0.014), (24, -0.121), (25, 0.13), (26, 0.05), (27, 0.069), (28, -0.079), (29, -0.12), (30, -0.054), (31, 0.019), (32, -0.026), (33, -0.079), (34, 0.002), (35, -0.017), (36, 0.045), (37, -0.033), (38, 0.082), (39, 0.065), (40, -0.005), (41, 0.061), (42, 0.01), (43, 0.044), (44, -0.067), (45, -0.048), (46, -0.001), (47, -0.041), (48, -0.038), (49, 0.038)]
simIndex simValue paperId paperTitle
same-paper 1 0.96735948 304 acl-2013-SEMILAR: The Semantic Similarity Toolkit
Author: Vasile Rus ; Mihai Lintean ; Rajendra Banjade ; Nobal Niraula ; Dan Stefanescu
Abstract: We present in this paper SEMILAR, the SEMantic simILARity toolkit. SEMILAR implements a number of algorithms for assessing the semantic similarity between two texts. It is available as a Java library and as a Java standalone ap-plication offering GUI-based access to the implemented semantic similarity methods. Furthermore, it offers facilities for manual se-mantic similarity annotation by experts through its component SEMILAT (a SEMantic simILarity Annotation Tool). 1
2 0.8826791 104 acl-2013-DKPro Similarity: An Open Source Framework for Text Similarity
Author: Daniel Bar ; Torsten Zesch ; Iryna Gurevych
Abstract: We present DKPro Similarity, an open source framework for text similarity. Our goal is to provide a comprehensive repository of text similarity measures which are implemented using standardized interfaces. DKPro Similarity comprises a wide variety of measures ranging from ones based on simple n-grams and common subsequences to high-dimensional vector comparisons and structural, stylistic, and phonetic measures. In order to promote the reproducibility of experimental results and to provide reliable, permanent experimental conditions for future studies, DKPro Similarity additionally comes with a set of full-featured experimental setups which can be run out-of-the-box and be used for future systems to built upon.
3 0.808801 12 acl-2013-A New Set of Norms for Semantic Relatedness Measures
Author: Sean Szumlanski ; Fernando Gomez ; Valerie K. Sims
Abstract: We have elicited human quantitative judgments of semantic relatedness for 122 pairs of nouns and compiled them into a new set of relatedness norms that we call Rel-122. Judgments from individual subjects in our study exhibit high average correlation to the resulting relatedness means (r = 0.77, σ = 0.09, N = 73), although not as high as Resnik’s (1995) upper bound for expected average human correlation to similarity means (r = 0.90). This suggests that human perceptions of relatedness are less strictly constrained than perceptions of similarity and establishes a clearer expectation for what constitutes human-like performance by a computational measure of semantic relatedness. We compare the results of several WordNet-based similarity and relatedness measures to our Rel-122 norms and demonstrate the limitations of WordNet for discovering general indications of semantic relatedness. We also offer a critique of the field’s reliance upon similarity norms to evaluate relatedness measures.
4 0.77202839 96 acl-2013-Creating Similarity: Lateral Thinking for Vertical Similarity Judgments
Author: Tony Veale ; Guofu Li
Abstract: Just as observing is more than just seeing, comparing is far more than mere matching. It takes understanding, and even inventiveness, to discern a useful basis for judging two ideas as similar in a particular context, especially when our perspective is shaped by an act of linguistic creativity such as metaphor, simile or analogy. Structured resources such as WordNet offer a convenient hierarchical means for converging on a common ground for comparison, but offer little support for the divergent thinking that is needed to creatively view one concept as another. We describe such a means here, by showing how the web can be used to harvest many divergent views for many familiar ideas. These lateral views complement the vertical views of WordNet, and support a system for idea exploration called Thesaurus Rex. We show also how Thesaurus Rex supports a novel, generative similarity measure for WordNet. 1 Seeing is Believing (and Creating) Similarity is a cognitive phenomenon that is both complex and subjective, yet for practical reasons it is often modeled as if it were simple and objective. This makes sense for the many situations where we want to align our similarity judgments with those of others, and thus focus on the same conventional properties that others are also likely to focus upon. This reliance on the consensus viewpoint explains why WordNet (Fellbaum, 1998) has proven so useful as a basis for computational measures of lexico-semantic similarity Guofu Li School of Computer Science and Informatics, University College Dublin, Belfield, Dublin D2, Ireland. l .guo fu . l gmai l i @ .com (e.g. see Pederson et al. 2004, Budanitsky & Hirst, 2006; Seco et al. 2006). These measures reduce the similarity of two lexical concepts to a single number, by viewing similarity as an objective estimate of the overlap in their salient qualities. This convenient perspective is poorly suited to creative or insightful comparisons, but it is sufficient for the many mundane comparisons we often perform in daily life, such as when we organize books or look for items in a supermarket. So if we do not know in which aisle to locate a given item (such as oatmeal), we may tacitly know how to locate a similar product (such as cornflakes) and orient ourselves accordingly. Yet there are occasions when the recognition of similarities spurs the creation of similarities, when the act of comparison spurs us to invent new ways of looking at an idea. By placing pop tarts in the breakfast aisle, food manufacturers encourage us to view them as a breakfast food that is not dissimilar to oatmeal or cornflakes. When ex-PM Tony Blair published his memoirs, a mischievous activist encouraged others to move his book from Biography to Fiction in bookshops, in the hope that buyers would see it in a new light. Whenever we use a novel metaphor to convey a non-obvious viewpoint on a topic, such as “cigarettes are time bombs”, the comparison may spur us to insight, to see aspects of the topic that make it more similar to the vehicle (see Ortony, 1979; Veale & Hao, 2007). In formal terms, assume agent A has an insight about concept X, and uses the metaphor X is a Y to also provoke this insight in agent B. To arrive at this insight for itself, B must intuit what X and Y have in common. But this commonality is surely more than a standard categorization of X, or else it would not count as an insight about X. To understand the metaphor, B must place X 660 Proce dingSsof oifa, th Beu 5l1gsarti Aan,An u aglu Mste 4e-ti9n2g 0 o1f3 t.he ?c A2s0s1o3ci Aatsiosonc fioartio Cno fmorpu Ctoamtiopnuatalt Lioin gauli Lsitnicgsu,i psatgices 6 0–670, in a new category, so that X can be seen as more similar to Y. Metaphors shape the way we per- ceive the world by re-shaping the way we make similarity judgments. So if we want to imbue computers with the ability to make and to understand creative metaphors, we must first give them the ability to look beyond the narrow viewpoints of conventional resources. Any measure that models similarity as an objective function of a conventional worldview employs a convergent thought process. Using WordNet, for instance, a similarity measure can vertically converge on a common superordinate category of both inputs, and generate a single numeric result based on their distance to, and the information content of, this common generalization. So to find the most conventional ways of seeing a lexical concept, one simply ascends a narrowing concept hierarchy, using a process de Bono (1970) calls vertical thinking. To find novel, non-obvious and useful ways of looking at a lexical concept, one must use what Guilford (1967) calls divergent thinking and what de Bono calls lateral thinking. These processes cut across familiar category boundaries, to simultaneously place a concept in many different categories so that we can see it in many different ways. de Bono argues that vertical thinking is selective while lateral thinking is generative. Whereas vertical thinking concerns itself with the “right” way or a single “best” way of looking at things, lateral thinking focuses on producing alternatives to the status quo. To be as useful for creative tasks as they are for conventional tasks, we need to re-imagine our computational similarity measures as generative rather than selective, expansive rather than reductive, divergent as well as convergent and lateral as well as vertical. Though WordNet is ideally structured to support vertical, convergent reasoning, its comprehensive nature means it can also be used as a solid foundation for building a more lateral and divergent model of similarity. Here we will use the web as a source of diverse perspectives on familiar ideas, to complement the conventional and often narrow views codified by WordNet. Section 2 provides a brief overview of past work in the area of similarity measurement, before section 3 describes a simple bootstrapping loop for acquiring richly diverse perspectives from the web for a wide variety of familiar ideas. These perspectives are used to enhance a Word- Net-based measure of lexico-semantic similarity in section 4, by broadening the range of informative viewpoints the measure can select from. Similarity is thus modeled as a process that is both generative and selective. This lateral-andvertical approach is evaluated in section 5, on the Miller & Charles (1991) data-set. A web app for the lateral exploration of diverse viewpoints, named Thesaurus Rex, is also presented, before closing remarks are offered in section 6. 2 Related Work and Ideas WordNet’s taxonomic organization of nounsenses and verb-senses – in which very general categories are successively divided into increasingly informative sub-categories or instancelevel ideas – allows us to gauge the overlap in information content, and thus of meaning, of two lexical concepts. We need only identify the deepest point in the taxonomy at which this content starts to diverge. This point of divergence is often called the LCS, or least common subsumer, of two concepts (Pederson et al., 2004). Since sub-categories add new properties to those they inherit from their parents – Aristotle called these properties the differentia that stop a category system from trivially collapsing into itself – the depth of a lexical concept in a taxonomy is an intuitive proxy for its information content. Wu & Palmer (1994) use the depth of a lexical concept in the WordNet hierarchy as such a proxy, and thereby estimate the similarity of two lexical concepts as twice the depth of their LCS divided by the sum of their individual depths. Leacock and Chodorow (1998) instead use the length of the shortest path between two concepts as a proxy for the conceptual distance between them. To connect any two ideas in a hierarchical system, one must vertically ascend the hierarchy from one concept, change direction at a potential LCS, and then descend the hierarchy to reach the second concept. (Aristotle was also first to suggest this approach in his Poetics). Leacock and Chodorow normalize the length of this path by dividing its size (in nodes) by twice the depth of the deepest concept in the hierarchy; the latter is an upper bound on the distance between any two concepts in the hierarchy. Negating the log of this normalized length yields a corresponding similarity score. While the role of an LCS is merely implied in Leacock and Chodorow’s use of a shortest path, the LCS is pivotal nonetheless, and like that of Wu & Palmer, the approach uses an essentially vertical reasoning process to identify a single “best” generalization. Depth is a convenient proxy for information content, but more nuanced proxies can yield 661 more rounded similarity measures. Resnick (1995) draws on information theory to define the information content of a lexical concept as the negative log likelihood of its occurrence in a corpus, either explicitly (via a direct mention) or by presupposition (via a mention of any of its sub-categories or instances). Since the likelihood of a general category occurring in a corpus is higher than that of any of its sub-categories or instances, such categories are more predictable, and less informative, than rarer categories whose occurrences are less predictable and thus more informative. The negative log likelihood of the most informative LCS of two lexical concepts offers a reliable estimate of the amount of infor- mation shared by those concepts, and thus a good estimate of their similarity. Lin (1998) combines the intuitions behind Resnick’s metric and that of Wu and Palmer to estimate the similarity of two lexical concepts as an information ratio: twice the information content of their LCS divided by the sum of their individual information contents. Jiang and Conrath (1997) consider the converse notion of dissimilarity, noting that two lexical concepts are dissimilar to the extent that each contains information that is not shared by the other. So if the information content of their most informative LCS is a good measure of what they do share, then the sum of their individual information contents, minus twice the content of their most informative LCS, is a reliable estimate of their dissimilarity. Seco et al. (2006) presents a minor innovation, showing how Resnick’s notion of information content can be calculated without the use of an external corpus. Rather, when using Resnick’s metric (or that of Lin, or Jiang and Conrath) for measuring the similarity of lexical concepts in WordNet, one can use the category structure of WordNet itself to estimate infor- mation content. Typically, the more general a concept, the more descendants it will possess. Seco et al. thus estimate the information content of a lexical concept as the log of the sum of all its unique descendants (both direct and indirect), divided by the log of the total number of concepts in the entire hierarchy. Not only is this intrinsic view of information content convenient to use, without recourse to an external corpus, Seco et al. show that it offers a better estimate of information content than its extrinsic, corpus-based alternatives, as measured relative to average human similarity ratings for the 30 word-pairs in the Miller & Charles (1991) test set. A similarity measure can draw on other sources of information besides WordNet’s category structures. One might eke out additional information from WordNet’s textual glosses, as in Lesk (1986), or use category structures other than those offered by WordNet. Looking beyond WordNet, entries in the online encyclopedia Wikipedia are not only connected by a dense topology of lateral links, they are also organized by a rich hierarchy of overlapping categories. Strube and Ponzetto (2006) show how Wikipedia can support a measure of similarity (and relatedness) that better approximates human judgments than many WordNet-based measures. Nonetheless, WordNet can be a valuable component of a hybrid measure, and Agirre et al. (2009) use an SVM (support vector machine) to combine information from WordNet with information harvested from the web. Their best similarity measure achieves a remarkable 0.93 correlation with human judgments on the Miller & Charles word-pair set. Similarity is not always applied to pairs of concepts; it is sometimes analogically applied to pairs of pairs of concepts, as in proportional analogies of the form A is to B as C is to D (e.g., hacks are to writers as mercenaries are to soldiers, or chisels are to sculptors as scalpels are to surgeons). In such analogies, one is really assessing the similarity of the unstated relationship between each pair of concepts: thus, mercenaries are soldiers whose allegiance is paid for, much as hacks are writers with income-driven loyalties; sculptors use chisels to carve stone, while surgeons use scalpels to cut or carve flesh. Veale (2004) used WordNet to assess the similarity of A:B to C:D as a function of the combined similarity of A to C and of B to D. In contrast, Turney (2005) used the web to pursue a more divergent course, to represent the tacit relationships of A to B and of C to D as points in a highdimensional space. The dimensions of this space initially correspond to linking phrases on the web, before these dimensions are significantly reduced using singular value decomposition. In the infamous SAT test, an analogy A:B::C:D has four other pairs of concepts that serve as likely distractors (e.g. singer:songwriter for hack:writer) and the goal is to choose the most appropriate C:D pair for a given A:B pairing. Using variants of Wu and Palmer (1994) on the 374 SAT analogies of Turney (2005), Veale (2004) reports a success rate of 38–44% using only WordNet-based similarity. In contrast, Turney (2005) reports up to 55% success on the same analogies, partly because his approach aims 662 to match implicit relations rather than explicit concepts, and in part because it uses a divergent process to gather from the web as rich a perspec- tive as it can on these latent relationships. 2.1 Clever Comparisons Create Similarity Each of these approaches to similarity is a user of information, rather than a creator, and each fails to capture how a creative comparison (such as a metaphor) can spur a listener to view a topic from an atypical perspective. Camac & Glucksberg (1984) provide experimental evidence for the claim that “metaphors do not use preexisting associations to achieve their effects [… ] people use metaphors to create new relations between concepts.” They also offer a salutary reminder of an often overlooked fact: every comparison exploits information, but each is also a source of new information in its own right. Thus, “this cola is acid” reveals a different perspective on cola (e.g. as a corrosive substance or an irritating food) than “this acid is cola” highlights for acid (such as e.g., a familiar substance) Veale & Keane (1994) model the role of similarity in realizing the long-term perlocutionary effect of an informative comparison. For example, to compare surgeons to butchers is to encourage one to see all surgeons as more bloody, … crude or careless. The reverse comparison, of butchers to surgeons, encourages one to see butchers as more skilled and precise. Veale & Keane present a network model of memory, called Sapper, in which activation can spread between related concepts, thus allowing one concept to prime the properties of a neighbor. To interpret an analogy, Sapper lays down new activation-carrying bridges in memory between analogical counterparts, such as between surgeon & butcher, flesh & meat, and scalpel & cleaver. Comparisons can thus have lasting effects on how Sapper sees the world, changing the pattern of activation that arises when it primes a concept. Veale (2003) adopts a similarly dynamic view of similarity in WordNet, showing how an analogical comparison can result in the automatic addition of new categories and relations to WordNet itself. Veale considers the problem of finding an analogical mapping between different parts of WordNet’s noun-sense hierarchy, such as between instances of Greek god and Norse god, or between the letters of different alphabets, such as of Greek and Hebrew. But no structural similarity measure for WordNet exhibits enough discernment to e.g. assign a higher similarity to Zeus & Odin (each is the supreme deity of its pantheon) than to a pairing of Zeus and any other Norse god, just as no structural measure will assign a higher similarity to Alpha & Aleph or to Beta & Beth than to any random letter pairing. A fine-grained category hierarchy permits fine-grained similarity judgments, and though WordNet is useful, its sense hierarchies are not especially fine-grained. However, we can automatically make WordNet subtler and more discerning, by adding new fine-grained categories to unite lexical concepts whose similarity is not reflected by any existing categories. Veale (2003) shows how a property that is found in the glosses of two lexical concepts, of the same depth, can be combined with their LCS to yield a new fine-grained parent category, so e.g. “supreme” + deity = Supreme-deity (for Odin, Zeus, Jupiter, etc.) and “1 st” + letter = 1st-letter (for Alpha, Aleph, etc.) Selected aspects of the textual similarity of two WordNet glosses – the key to similarity in Lesk (1986) – can thus be reified into an explicitly categorical WordNet form. 3 Divergent (Re)Categorization To tap into a richer source of concept properties than WordNet’s glosses, we can use web ngrams. Consider these descriptions of a cowboy from the Google n-grams (Brants & Franz, 2006). The numbers to the right are Google frequency counts. a lonesome cowboy 432 a mounted cowboy 122 a grizzled cowboy 74 a swaggering cowboy 68 To find the stable properties that can underpin a meaningful fine-grained category for cowboy, we must seek out the properties that are so often presupposed to be salient of all cowboys that one can use them to anchor a simile, such as
5 0.72878933 43 acl-2013-Align, Disambiguate and Walk: A Unified Approach for Measuring Semantic Similarity
Author: Mohammad Taher Pilehvar ; David Jurgens ; Roberto Navigli
Abstract: Semantic similarity is an essential component of many Natural Language Processing applications. However, prior methods for computing semantic similarity often operate at different levels, e.g., single words or entire documents, which requires adapting the method for each data type. We present a unified approach to semantic similarity that operates at multiple levels, all the way from comparing word senses to comparing text documents. Our method leverages a common probabilistic representation over word senses in order to compare different types of linguistic data. This unified representation shows state-ofthe-art performance on three tasks: seman- tic textual similarity, word similarity, and word sense coarsening.
6 0.70176148 262 acl-2013-Offspring from Reproduction Problems: What Replication Failure Teaches Us
7 0.65886742 222 acl-2013-Learning Semantic Textual Similarity with Structural Representations
8 0.65333205 185 acl-2013-Identifying Bad Semantic Neighbors for Improving Distributional Thesauri
9 0.64077693 31 acl-2013-A corpus-based evaluation method for Distributional Semantic Models
10 0.59549022 27 acl-2013-A Two Level Model for Context Sensitive Inference Rules
11 0.57005924 93 acl-2013-Context Vector Disambiguation for Bilingual Lexicon Extraction from Comparable Corpora
12 0.56041288 268 acl-2013-PATHS: A System for Accessing Cultural Heritage Collections
13 0.55433995 242 acl-2013-Mining Equivalent Relations from Linked Data
14 0.55112284 281 acl-2013-Post-Retrieval Clustering Using Third-Order Similarity Measures
15 0.54154807 376 acl-2013-Using Lexical Expansion to Learn Inference Rules from Sparse Data
16 0.53802836 291 acl-2013-Question Answering Using Enhanced Lexical Semantic Models
17 0.53182071 284 acl-2013-Probabilistic Sense Sentiment Similarity through Hidden Emotions
18 0.51844335 238 acl-2013-Measuring semantic content in distributional vectors
19 0.51672208 231 acl-2013-Linggle: a Web-scale Linguistic Search Engine for Words in Context
20 0.51268113 182 acl-2013-High-quality Training Data Selection using Latent Topics for Graph-based Semi-supervised Learning
topicId topicWeight
[(0, 0.089), (6, 0.036), (11, 0.054), (15, 0.303), (24, 0.07), (26, 0.034), (28, 0.012), (35, 0.09), (42, 0.046), (48, 0.038), (64, 0.013), (70, 0.043), (88, 0.024), (90, 0.024), (95, 0.055)]
simIndex simValue paperId paperTitle
1 0.96840543 232 acl-2013-Linguistic Models for Analyzing and Detecting Biased Language
Author: Marta Recasens ; Cristian Danescu-Niculescu-Mizil ; Dan Jurafsky
Abstract: Unbiased language is a requirement for reference sources like encyclopedias and scientific texts. Bias is, nonetheless, ubiquitous, making it crucial to understand its nature and linguistic realization and hence detect bias automatically. To this end we analyze real instances of human edits designed to remove bias from Wikipedia articles. The analysis uncovers two classes of bias: framing bias, such as praising or perspective-specific words, which we link to the literature on subjectivity; and epistemological bias, related to whether propositions that are presupposed or entailed in the text are uncontroversially accepted as true. We identify common linguistic cues for these classes, including factive verbs, implicatives, hedges, and subjective inten- cs . sifiers. These insights help us develop features for a model to solve a new prediction task of practical importance: given a biased sentence, identify the bias-inducing word. Our linguistically-informed model performs almost as well as humans tested on the same task.
2 0.82956612 262 acl-2013-Offspring from Reproduction Problems: What Replication Failure Teaches Us
Author: Antske Fokkens ; Marieke van Erp ; Marten Postma ; Ted Pedersen ; Piek Vossen ; Nuno Freire
Abstract: Repeating experiments is an important instrument in the scientific toolbox to validate previous work and build upon existing work. We present two concrete use cases involving key techniques in the NLP domain for which we show that reproducing results is still difficult. We show that the deviation that can be found in reproduction efforts leads to questions about how our results should be interpreted. Moreover, investigating these deviations provides new insights and a deeper understanding of the examined techniques. We identify five aspects that can influence the outcomes of experiments that are typically not addressed in research papers. Our use cases show that these aspects may change the answer to research questions leading us to conclude that more care should be taken in interpreting our results and more research involving systematic testing of methods is required in our field.
3 0.82142818 199 acl-2013-Integrating Multiple Dependency Corpora for Inducing Wide-coverage Japanese CCG Resources
Author: Sumire Uematsu ; Takuya Matsuzaki ; Hiroki Hanaoka ; Yusuke Miyao ; Hideki Mima
Abstract: This paper describes a method of inducing wide-coverage CCG resources for Japanese. While deep parsers with corpusinduced grammars have been emerging for some languages, those for Japanese have not been widely studied, mainly because most Japanese syntactic resources are dependency-based. Our method first integrates multiple dependency-based corpora into phrase structure trees and then converts the trees into CCG derivations. The method is empirically evaluated in terms of the coverage of the obtained lexi- con and the accuracy of parsing.
same-paper 4 0.79914433 304 acl-2013-SEMILAR: The Semantic Similarity Toolkit
Author: Vasile Rus ; Mihai Lintean ; Rajendra Banjade ; Nobal Niraula ; Dan Stefanescu
Abstract: We present in this paper SEMILAR, the SEMantic simILARity toolkit. SEMILAR implements a number of algorithms for assessing the semantic similarity between two texts. It is available as a Java library and as a Java standalone ap-plication offering GUI-based access to the implemented semantic similarity methods. Furthermore, it offers facilities for manual se-mantic similarity annotation by experts through its component SEMILAT (a SEMantic simILarity Annotation Tool). 1
5 0.75158006 233 acl-2013-Linking Tweets to News: A Framework to Enrich Short Text Data in Social Media
Author: Weiwei Guo ; Hao Li ; Heng Ji ; Mona Diab
Abstract: Many current Natural Language Processing [NLP] techniques work well assuming a large context of text as input data. However they become ineffective when applied to short texts such as Twitter feeds. To overcome the issue, we want to find a related newswire document to a given tweet to provide contextual support for NLP tasks. This requires robust modeling and understanding of the semantics of short texts. The contribution of the paper is two-fold: 1. we introduce the Linking-Tweets-toNews task as well as a dataset of linked tweet-news pairs, which can benefit many NLP applications; 2. in contrast to previ- ous research which focuses on lexical features within the short texts (text-to-word information), we propose a graph based latent variable model that models the inter short text correlations (text-to-text information). This is motivated by the observation that a tweet usually only covers one aspect of an event. We show that using tweet specific feature (hashtag) and news specific feature (named entities) as well as temporal constraints, we are able to extract text-to-text correlations, and thus completes the semantic picture of a short text. Our experiments show significant improvement of our new model over baselines with three evaluation metrics in the new task.
6 0.73937684 47 acl-2013-An Information Theoretic Approach to Bilingual Word Clustering
7 0.58279371 187 acl-2013-Identifying Opinion Subgroups in Arabic Online Discussions
8 0.57647932 250 acl-2013-Models of Translation Competitions
9 0.56738693 318 acl-2013-Sentiment Relevance
10 0.56709927 83 acl-2013-Collective Annotation of Linguistic Resources: Basic Principles and a Formal Model
11 0.56591165 99 acl-2013-Crowd Prefers the Middle Path: A New IAA Metric for Crowdsourcing Reveals Turker Biases in Query Segmentation
12 0.56225979 346 acl-2013-The Impact of Topic Bias on Quality Flaw Prediction in Wikipedia
13 0.55938894 59 acl-2013-Automated Pyramid Scoring of Summaries using Distributional Semantics
14 0.55858642 2 acl-2013-A Bayesian Model for Joint Unsupervised Induction of Sentiment, Aspect and Discourse Representations
15 0.5553689 224 acl-2013-Learning to Extract International Relations from Political Context
16 0.55436498 63 acl-2013-Automatic detection of deception in child-produced speech using syntactic complexity features
17 0.55205053 17 acl-2013-A Random Walk Approach to Selectional Preferences Based on Preference Ranking and Propagation
18 0.54520506 43 acl-2013-Align, Disambiguate and Walk: A Unified Approach for Measuring Semantic Similarity
19 0.54046911 230 acl-2013-Lightly Supervised Learning of Procedural Dialog Systems
20 0.53855598 52 acl-2013-Annotating named entities in clinical text by combining pre-annotation and active learning