Adapting a Constituency Parser to User-Generated Content in Polish Opinion Mining

Pluwak, Agnieszka; Korczyński, Wojciech; Kisiel-Dorohinicki, Marek

doi:10.7494/csci.2016.17.1.23

Cited by 2 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the nature of this kind of data, the texts commonly include many mistakes and problematic phenomena. A linguistic analysis of the data as well as the results of similar research (see e.g (Pluwak et al, 2016)) helped us to define distinctive features, which are related to various text levels such as: notation (e.g. lack of diacritics, spelling mistakes, typos, omissions of capital letters, incorrectly connected or disconnected segments, lack of or poor punctuation), morphology and syntax (e.g.…”

Section: Computer-mediated Communication Corpusmentioning

confidence: 99%

Tagger for Polish Computer Mediated Communication Texts

Walentynowicz

Piasecki

Oleksy

2019

Proceedings - Natural Language Processing in a Deep Learning World

View full text Add to dashboard Cite

In this paper we present a morphosyntactic tagger dedicated to Computermediated Communication texts in Polish. Its construction is based on an expanded RNN-based neural network adapted to the work on noisy texts. Among several techniques, the tagger utilises fastText embedding vectors, sequential character embedding vectors, and Brown clustering for the coarse-grained representation of sentence structures. In addition a set of manually written rules was proposed for postprocessing. The system was trained to disambiguate descriptions of words in relation to Parts of Speech tags together with the full morphological information in terms of values for the different grammatical categories. We present also evaluation of several model variants on the gold standard annotated CMC data, comparison to the state-of-the-art taggers for Polish and error analysis. The proposed tagger shows significantly better results in this domain and demonstrates the viability of adaptation.

show abstract

Section: Computer-mediated Communication Corpusmentioning

confidence: 99%

Tagger for Polish Computer Mediated Communication Texts

Walentynowicz

Piasecki

Oleksy

2019

Proceedings - Natural Language Processing in a Deep Learning World

View full text Add to dashboard Cite

show abstract

Towards Application of Speech Act Theory to Opinion Mining

Pluwak

2016

View full text Add to dashboard Cite

Towards the Application of Speech Act Theory to Opinion MiningThe paper refers to the pragmatics’ perspective on opinion mining in Polish and English, inspired by the discrepancy between the coverage of sentiment analysis and the market demand. An analysis of speech acts expressed in opinion texts reveals that almost half of all opinions include ways of indirect evaluation that might not get extracted while applying traditional methods of sentiment analysis based on direct evaluative vocabulary and polarity lexicons. Coding of sentiment with respect to speech acts could vastly broaden data mining results within an NLP-system. O zastosowaniu teorii aktów mowy w ekstrakcji danych z tekstów opinii internetowychJedno z aktualnych zagadnień językoznawstwa komputerowego, jakim jest automatyczne badanie wydźwięku wypowiedzi, nie uwzględniło dotychczas w wystarczającym stopniu pragmatyki językoznawczej, np. aktów mowy Austina (1961) i Searla (1969), a zatem również implicytnych sposobów wyrażania ewaluacji. Tymczasem podejście od pragmatyki ku konstrukcjom przełożonym na reguły programistyczne umożliwiłoby nie tylko szersze spojrzenie na analizę sentymentu, ale też zbliżyłoby automat do sposobu, w jaki odbiera go człowiek. W szczególności chodzi tu sposoby wyrażania (nie)zadowolenia wykraczające poza poziom leksykalny (bez nacechowanej negatywnie leksyki), typu Nigdy więcej tam nie pójdę.Artykuł prezentuje: 1. aktualne podejścia do analizy wydźwięku w lingwistyce komputerowej, 2. propozycję zastosowania podejścia pragmatycznego, 3. wyniki badania próbki tekstów opinii internetowych pod kątem występowania w nich aktów mowy, 4. propozycję utworzenia reguł ekstrakcji danych na ich podstawie. Zaprezentowane podejście zakłada hipotezę wtórnej oralności, czyli tego, że język opinii jest zapisanym językiem mówionym.

show abstract

Adapting a Constituency Parser to User-Generated Content in Polish Opinion Mining

Cited by 2 publications

References 14 publications

Tagger for Polish Computer Mediated Communication Texts

Tagger for Polish Computer Mediated Communication Texts

Towards Application of Speech Act Theory to Opinion Mining

Contact Info

Product

Resources

About