Then we propose various novel similarity measurements including surface features, meta-path based semantic features and social correlation features and combine them in a learning-to-rank framework.

0 We propose two new similarity measures , as well as integrating temporal information into

the similarity measures to generate global semantic features.

We first extract surface features between the morph and the candidate based on measuring orthographic similarity measures which were commonly used in entity coreference resolution (e.g.

4.2.3 Meta-Path-Based Semantic Similarity Measurements

We then adopt meta-path-based similarity measures (Sun et al., 2011a; Sun et al., 2011b), which are defined over heterogeneous networks to extract semantic features.

where sim(v, v’) is a vector similarity measure .

We note that the general DIRT scheme may be used while employing other “base” vector similarity measures .

This issue has been addressed in a separate line of research which introduced directional similarity measures suitable for inference relations (Bhagat et al., 2007; Szpektor and Dagan, 2008; Kotlerman et al., 2010).

Our scheme can be applied on top of any context-insensitive “base” similarity measure for rule learning, which operates at the word level, such as Cosine or Lin (Lin, 1998).

We apply our two-level scheme over three state-of-the-art context-insensitive similarity measures .

To apply our PSSS on IQAPS inference task, we use it as the sentiment similarity measure in the algorithm explained in Figure 4.

The second row of Table 4 show the results of using a popular semantic similarity measure , PMI, as the sentiment similarity (SS) measure in Figure 4.

Table 4 shows the effectiveness of our sentiment similarity measure .

Semantic similarity measures such as Latent Semantic Analysis (LSA) (Landauer et al., 1998) can effectively capture the similarity between semantically related words like "car" and "automobile", but they are less effective in relating words with similar sentiment orientation like "excellent" and "superior".

We show that our approach effectively outperforms the semantic similarity measures in two NLP tasks: Indirect yes/no Question Answer Pairs (IQAPs) Inference and Sentiment Orientation (S0) prediction that are described as follows:

Previous research utilized the semantic relations between words obtained from WordNet (Hassan and Radev, 2010) and semantic similarity measures (e.g.

They also utilized Latent Semantic Analysis (LSA) (Landauer et al., 1998) as another semantic similarity measure .

However, both PM and LSA are semantic similarity measure .

(2012), we used two semantic similarity measures (PMI and LSA) for the IQAP inference task.

As we discussed above, semantic similarity measures are less effective to infer sentiment similarity between word pairs.

Within this context, we propose a new methodology that adapts the classical K -means algorithm to a third-order similarity measure initially developed for NLP tasks.

In this paper, we proposed a new PRC approach which (1) is based on the adaptation of the K -means algorithm to third-order similarity measures and (2) proposes a coherent stopping criterion.

In particular, p is the size of the word feature vectors representing both Web snippets and centroids (p = 2.5), K is the number of clusters to be found (K = 2..10) and 8(Wik, 14/31) is the collocation measure integrated in the InfoSimba similarity measure .

feature vectors are hard to define in small collections of short text fragments (Timonen, 2013), (2) existing second-order similarity measures such as the cosine are unadapted to capture the semantic similarity between small texts, (3) Latent Semantic Analysis has evidenced inconclusive results (Osinski and Weiss, 2005) and (4) the labeling process is a surprisingly hard extra task (Carpineto et al., 2009).

For that purpose, we propose a new methodology that adapts the classical K -means algorithm to a third-order similarity measure initially developed for Topic Segmentation (Dias et al., 2007).

Within the context of PRC, the K -means algorithm needs to be adapted to integrate third-order similarity measures (Mihalcea et al., 2006; Dias et al., 2007).

Third-order similarity measures, also called weighted second-order similarity measures, do not rely on exact matches of word features as classical second-order similarity measures (6. g. the cosine metric), but rather evaluate similarity based on related matches.

In this paper, we propose to use the third-order similarity measure called InfoSimba introduced in (Dias et al., 2007) for Topic Segmentation and implement its simplified version 83 in Equation 2.

In order to compare semantic signatures, we adopt the Cosine similarity measure as a baseline method.

The top-ranking participating systems in the SemEval-2012 task were generally supervised systems utilizing a variety of lexical resources and similarity measurement techniques.

3.3 Similarity Measure Analysis

In addition, we present in the table correlation scores for four other similarity measures reported by B'ar et al.

Different evaluation methods exist in the literature for evaluating the performance of a word-level semantic similarity measure ; we adopted two well-established benchmarks: synonym recognition and correlating word similarity judgments with those from human annotators.

We adopt this task as a way of evaluating our similarity measure at the sense level.

For each of these paraphrases, a DP is constructed and compared to that of the oov word using a similarity measure (Section 2.2).

where t is a phrase on the target side, 0 is the oov word or phrase, and s is a paraphrase of 0. p(s|0) is estimated using a similarity measure over DPs and p(t|s) is coming from the phrase-table.

2.3 Similarity Measures

In Section 2.2 and 2.3, different types of association measures and similarity measures have been explained to build and compare distributional profiles.

As the results show, the combination of PMI as association measure and cosine as DP similarity measure outperforms the other possible combinations.

Each phrase type represents a vertex in the graph and is connected to other vertices with a weight defined by a similarity measure between the two profiles (Section 2.3).

However based on the definition of the similarity measures using context, it is quite possible that an oov node and a labeled node which are connected to the same unlabeled node do not share any context words and hence are not directly connected.

In such a graph, the similarity of each pair of nodes is computed using one of the similarity measures discussed above.

As a consequence, improving such thesaurus is an important issue that is mainly tackled indirectly through the improvement of semantic similarity measures .

As in (Lin, 1998) or (Curran and Moens, 2002a), this building is based on the definition of a semantic similarity measure from a corpus.

For the extraction of distributional data and the characteristics of the distributional similarity measure , we adopted the options of (Ferret, 2010), resulting from a kind of grid search procedure performed with the extended TOEFL test proposed in (Freitag et al., 2005) as an optimization objective.

o similarity measure between contexts, for evaluating the semantic similarity of two words = Cosine measure.

Following work such as (Grefenstette, 1994), a widespread way to build a thesaurus from a corpus is to use a semantic similarity measure for extracting the semantic neighbors of the entries of the thesaurus.

Work based on WordNet-like lexical networks for building semantic similarity measures such as (Budanitsky and Hirst, 2006) or (Pedersen et al., 2004) falls into this category.

A part of these proposals focus on the weighting of the elements that are part of the contexts of words such as (Broda et al., 2009), in which the weights of context elements are turned into ranks, or (Zhitomirsky-Geffet and Dagan, 2009), followed and extended by (Yamamoto and Asakura, 2010), that proposes a bootstrapping method for modifying the weights of context elements according to the semantic neighbors found by an initial distributional similarity measure .

For instance, features such as ngrams of words or ngrams of parts of speech are not considered whereas they are widely used in tasks such as word sense disambiguation (WSD) for instance, probably because they would lead to very large models and because similarity measures such as the Cosine measure are not necessarily suitable for heterogeneous representations (Alexandrescu and Kirchhoff, 2007).

As a consequence, the improvement of such thesaurus is generally not directly addressed but is a possible consequence of the improvement of semantic similarity measures .

We show also how Thesaurus Rex supports a novel, generative similarity measure for WordNet.

more rounded similarity measures .

A similarity measure can draw on other

Their best similarity measure achieves a remarkable 0.93 correlation with human judgments on the Miller & Charles word-pair set.

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.

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.

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.

Instead of utilizing simple similarity measures and their disjoint combinations, our method directly optimizes document and entity representations for a given similarity measure .

A supervised fine-tuning stage follows to optimize the representation towards the similarity measure .

We propose a deep learning approach that automatically learns context-entity similarity measure for entity disambiguation.

When embedding our similarity measure sim(d, 6) into (Han et al., 2011), we achieve the best results on AIDA.

ument and entity representations for a fixed similarity measure .

In fact, the underlying representations for computing similarity measure add internal structure to the given similarity measure .

The learned similarity measure can be readily incorporated into any existing collective approaches, which further boosts performance.

In this work, we analyze various Japanese corpora using a number of collocation and word similarity measures to deduce and suggest the best collocations for Japanese second language learners.

In order to build a system that is more sensitive to constructions that are difficult for learners, we use word similarity measures that generate collocation candidates using a large Japanese language learner corpus.

similarity measures are used.

Our work follows the general approach, that is, uses similarity measures for generating the confusion set and association measures for ranking the best candidates.

Similarity measures are used to generate the collocation candidates that are later ranked using association measures.

They align senses in WordNet to Wikipedia entries in a supervised setting using semantic similarity measures .

Niemann and Gurevych (2011) combine two different types of similarity (i) cosine similarity on bag-of-words vectors (COS) and (ii) a personalized PageRank—based similarity measure (PPR).

For each similarity measure , Niemann and Gurevych (2011) determine a threshold (tppr and

The similarity measure is based on stem overlap of the candidates’ glosses expanded by WordNet domains, the WordNet synset, and the set of senses for a FrameNet frame.