Opinion Analysis Across Languages: An Overview of and Observations from the NTCIR6 Opinion Analysis Pilot Task

Evans, David K.; Ku, Lun-Wei; Seki, Yohei; Chen, Hsin-Hsi; Kando, Noriko

doi:10.1007/978-3-540-73400-0_57

Cited by 15 publications

(10 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Five teams participated in OAT-6 evaluation [15]. The achieved F 1 s on opinionated sentence identification range from 0.331 to 0.448 under strict Learning Multiple Level Features for Opinion Analysis 363 evaluation, and from 0.625 to 0.776 under lenient evaluation, respectively.…”

Section: Reference and Baseline Systemsmentioning

confidence: 99%

“…A systematic observation and analysis is performed on NTCIR-6 opinion corpus training set (OAT-6 train) to discover the linguistic clues for opinion analysis [15]. This dataset consists of 429 news documents in Chinese corresponding to 4 topics.…”

Section: Linguistic Clues Discovery For Opinion Analysismentioning

confidence: 99%

“…In OAT-6 and MOAT-7 testing dataset, all of the sentences were annotated by three assessors. A strict (all three assessors must have the same annotation) and a lenient standard (two of three assessors have the same annotation) were generated, respectively [15,34]. OAT-6 consists of 11,907 sentences in Traditional Chinese in which 25% and 62% ones are opinionated under strict and lenient standard, respectively.…”

Section: Experiments Setupmentioning

confidence: 99%

“…Based on the systematic observation and analysis on the Chinese training part of NTCIR-6 opinion corpus (OAT-6 Train) [15], multiple level linguistic clues and corresponding features are discovered, including: (1) punctuations; (2) the wordlevel features such as opinion words/emotion words/scene words, the opinion operators and opinion indicators; (3) the collocation-level features such as the collocations between opinion holders and opinion operators, the collocations between degree adverbs and sentiment words, and especially the collocation behaviors of opinion words and objects; and (4) the inter-sentence opinion orientation transformation behaviors. Accordingly, a set of opinion-related lexicons are complied.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Learning Multiple Level Features for Opinion Analysis

Kit

2011

Int. J. Comp. Proc. Lang.

View full text Add to dashboard Cite

In recent years, there has been an increasing interest in opinion analysis which seeks to identify and classify opinions in the text automatically. Many existing opinion analysis techniques use opinion lexicon as the most important features but lack the learning capacity of new opinion words and especially, the contextual behaviors of opinion words. Additionally, many works employ intra-sentence features while the contextual intersentence level feature is not well studied. This paper presents an opinion analysis technique which incorporates multiple level intra-sentence features from punctuation-, word-to collocation-level and inter-sentence feature. Based on the systematic observation and analysis on the opinion corpus, some linguistic clues and corresponding features for opinion analysis are discovered. These features are incorporated in a support vector machine (SVM) based classifier to identify opinionated sentences from running text and determine their polarities. To overcome the barrier caused by insufficient annotated training data, a semi-supervised learning algorithm is designed to train the classifier and enrich the opinion-related knowledge by using high quality instances identified from large raw text as new training data interactively. Evaluations on the datasets of NTCIR-6 opinion analysis task (OAT-6) and NTCIR-7 multilingual opinion analysis task (MOAT-7) show that the proposed opinion analysis approach achieved encouraging performance.

show abstract

Section: Reference and Baseline Systemsmentioning

confidence: 99%

Section: Linguistic Clues Discovery For Opinion Analysismentioning

confidence: 99%

Section: Experiments Setupmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Learning Multiple Level Features for Opinion Analysis

Kit

2011

Int. J. Comp. Proc. Lang.

View full text Add to dashboard Cite

show abstract

“…Evans et al [3] create such corpora by annotating English, Chinese and Japanese texts on a sentence level.…”

Section: Related Workmentioning

confidence: 99%

Generic high-throughput methods for multilingual sentiment detection

Gindl

Scharl

Weichselbraun

2010

4th IEEE International Conference on Digital Ecosystems and Technologies

View full text Add to dashboard Cite

Abstract-Digital ecosystems typically involve a large number of participants from different sectors who generate rapidly growing archives of unstructured text. Measuring the frequency of certain terms to determine the popularity of a topic is comparably straightforward. Detecting sentiment expressed in user-generated electronic content is more challenging, especially in the case of digital ecosystems comprising heterogeneous sets of multilingual documents. This paper describes the use of language-specific grammar patterns and multilingual tagged dictionaries to detect sentiment in German and English document repositories. Digital ecosystems may contain millions of frequently updated documents, requiring sentiment detection methods that maximize throughput. The ideal combination of high-throughput techniques and more accurate (but slower) approaches depends on the specific requirements of an application. To accommodate a wide range of possible applications, this paper presents (i) an adaptive method, balancing accuracy and scalability of multilingual textual sources, (ii) a generic approach for generating language-specific grammar patterns and multilingual tagged dictionaries, and (iii) an extensive evaluation verifying the method's performance based on Amazon product reviews and user evaluations from Sentiment Quiz, a game-with-a-purpose that invites users of the Facebook social networking platform to assess the sentiment of individual sentences.

show abstract