Multilingual Sentiment Analysis for Enhanced Language Resources

Slide Note
Embed
Share

This presentation discusses the EUROSENTIMENT project focusing on generating multilingual variants for sentiment lexicons. Addressing challenges in sentiment analysis, the project aims to build a shared language resource pool to improve adaptability and interoperability of language resources. Objectives include reducing costs, providing semantic interoperability, and demonstrating the impact on multilingual sentiment analysis applications.

tram_isa
tram_isa Follow

Uploaded on Dec 09, 2024 | 0 Views

Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

E N D

Presentation Transcript

  1. Generating Multilingual Variants for Automatically Extracted Sentiment Lexicons The EUROSENTIMENT use-case Gabriela Vulcu, Paul Buitelaar Insight, Centre for Data Analytics, National University of Ireland, Galway 09/12/2024 Presenter name 1

  2. Outline 1. EUROSENTIMENT overview 2. Legacy Language Resources 3. Pipeline for Language Resource Adaptation 4. Tools and Demos EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 2

  3. EUROSENTIMENT: problems addressed Overall problems lack of agreed schemas for sentiment analysis (e.g. no agreed sentiment strength) lack of visibility, accessibility of sentiment analysis language resources Addressed problems high costs for adapting sentiment analysis resources to new domains, languages lack of interoperability with other language or semantic resources like the Linguistic Linked Open Data cloud (LLOD) EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 3

  4. EUROSENTIMENT: objectives Overall objectives Create a market for language resources by setting up a shared language resource pool (LRP) Putting in place best-practice guidelines, QA procedures for the LRP Demonstrate the impact of the LRP in multilingual sentiment analysis applications Addressed objectives Reduce the cost of creating, adapting sentiment language resources by using the LRP Provide semantic interoperability between multilingual sentiment analysis language resources EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 4

  5. Outline 1. EUROSENTIMENT overview 2. Legacy Language Resources 3. Pipeline for Language Resource Adaptation 4. Tools and Demos EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 5

  6. Heterogeneity of Existing Language Resources Format plain text HTML, XML, EXCEL, TSV, CSV, RDF/XML with or without custom made annotations, Annotations domain, language entities, lemma, POS, WordNet synset sentiment emotion inflections EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 6

  7. Example: TripAdvisor hotels dataset <Author>kekeScotland <Content>We got a good deal compared to other Madrid prices and as we were only using the hotel as a base so we didn't expect too much. Location of the Alberto Aguilera NH hotel was pretty good and the room was quiet and clean. We had a safe and free wi-fi. Only downside was the room was on the small side and the bathroom was tiny. <Date>Sep 10, 2008 <Value>3 <Rooms>4 <No. Reader>-1 <Location>5 <No. Helpful>-1 <Cleanliness>4 <Overall>4 <Check in / front desk>1 <Service>3 <Business service>4 EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 7

  8. Polarity of specific aspects <Author>kekeScotland <Content>We got a good deal compared to other Madrid prices and as we were only using the hotel as a base so we didn't expect too much. Location of the Alberto Aguilera NH hotel was pretty good and the room was quiet and clean. We had a safe and free wi-fi. Only downside was the room was on the small side and the bathroom was tiny. <Date>Sep 10, 2008 <Value>3 <Rooms>4 <No. Reader>-1 <Location>5 <No. Helpful>-1 <Cleanliness>4 <Overall>4 <Check in / front desk>1 <Service>3 <Business service>4 EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 8

  9. Example: a different format qmazon electronics dataset [ t ] the best 4mp compact digital available camera[+2]## this camera is perfect for an enthusiastic amateur photographer . picture[+3], macro[+3]## the pictures are razor-sharp , even in macro . . . EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 9

  10. Aspect polarity and entities annotations Amazon electronics dataset [ t ] the best 4mp compact digital available camera[+2]## this camera is perfect for an enthusiastic amateur photographer . picture[+3], macro[+3]## the pictures are razor-sharp , even in macro . . . EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 10

  11. Proprietary formats Paradigma dataset 0 :place::NNS;;that::that::WDT;;are::be::VBP;;far::far::RB;;bett er::good::JJR;;,::,::Fc;;but::but::CC;;compared::compare::VB N;;to::to::TO;;Pizza::pizza::NN;;Hut::hut::NN;;or::or::CC;;Do minos::domino::NNS;;it::it::PRP;;wins::win::VBZ;;.::.::Fp positive 10 There::there::EX;;are::be::VBP;;local::local::JJ;;places: en PIZZA HUT/PIZZA HUT EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 11

  12. Lemma, POS, overall polarity Paradigma dataset 0 :place::NNS;;that::that::WDT;;are::be::VBP;;far::far::RB;;bett er::good::JJR;;,::,::Fc;;but::but::CC;;compared::compare::VB N;;to::to::TO;;Pizza::pizza::NN;;Hut::hut::NN;;or::or::CC;;Do minos::domino::NNS;;it::it::PRP;;wins::win::VBZ;;.::.::Fp positive 10 There::there::EX;;are::be::VBP;;local::local::JJ;;places: en PIZZA HUT/PIZZA HUT EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 12

  13. Outline 1. EUROSENTIMENT overview 2. Legacy Language Resources 3. Pipeline for Language Resource Adaptation 4. Tools and Demos EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 13

  14. Pipeline for Language Resource Adaptation domain-specific sentiment lexicon generation entity extraction for aspect-based sentiment analysis sentiment analysis in the context of entities lexicons are modeled using lemon sentiments are modeled using Marl EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 14

  15. Corpus Conversion adapts sentiment corpora to a common schema (NIF, Marl) EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 15

  16. Example: Corpus Conversion "entries": [ { "nif:String": We got a good deal compared to other Madrid prices and as we were only using the hotel as a base so we didn't expect too much. Location of the Alberto Aguilera NH hotel was pretty good and the room was quiet and clean. We had a safe and free wi-fi. Only downside was the room was on the small side and the bathroom was tiny.", "dc:language": "en", "prov:wasDerivedFrom": { @type": "es:TripadvisorComment , "date": "2008-09-10", "user": "kekeScotland }, "opinions": [ { "@id": "_:Opinion01", "marl:hasPolarity": "marl:Positive", "marl:polarityValue": 4, "marl:describesObjectFeature": "Overall }, }] }] EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 16

  17. Example: Corpus Conversion "entries": [ { "nif:String": We got a good deal compared to other Madrid prices and as we were only using the hotel as a base so we didn't expect too much. Location of the Alberto Aguilera NH hotel was pretty good and the room was quiet and clean. We had a safe and free wi-fi. Only downside was the room was on the small side and the bathroom was tiny.", "dc:language": "en", "prov:wasDerivedFrom": { @type": "es:TripadvisorComment , "date": "2008-09-10", "user": "kekeScotland }, "opinions": [ { "@id": "_:Opinion01", "marl:hasPolarity": "marl:Positive", "marl:polarityValue": 4, "marl:describesObjectFeature": "Overall }, }] }] EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 17

  18. Example: Corpus Conversion "entries": [ { "nif:String": We got a good deal compared to other Madrid prices and as we were only using the hotel as a base so we didn't expect too much. Location of the Alberto Aguilera NH hotel was pretty good and the room was quiet and clean. We had a safe and free wi-fi. Only downside was the room was on the small side and the bathroom was tiny.", "dc:language": "en", "prov:wasDerivedFrom": { @type": "es:TripadvisorComment , "date": "2008-09-10", "user": "kekeScotland }, "opinions": [ { "@id": "_:Opinion01", "marl:hasPolarity": "marl:Positive", "marl:polarityValue": 4, "marl:describesObjectFeature": "Overall }, }] }] EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 18

  19. Semantic Analysis entity class extraction (hotel) named entity extraction (Alberto Aguilera NH) linking to LOD (DBpedia) and LLOD cloud (WordNet, BabelNet) EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 19

  20. Example: Entities and Entity Classes Extraction We got a good deal compared to other Madrid prices and as we were only using the hotel as a base so we didn't expect too much. Location of the Alberto Aguilera NH hotel was pretty good and the room was quiet and clean. We had a safe and free wi-fi. Only downside was the room was on the small side and the bathroom was tiny. " Entity class Dbpedia link WordNet link deal http://dbpedia.org/resource/Deal http://wordnet-rdf.princeton.edu/wn31/1110274-n location http://wordnet-rdf.princeton.edu/wn31/27167-n http://dbpedia.org/resource/Location_(geograp hy) hotel http://dbpedia.org/resource/Hotel http://wordnet-rdf.princeton.edu/wn31/3542333-n room http://dbpedia.org/resource/Room http://wordnet-rdf.princeton.edu/wn31/4105893-n wi-fi http://dbpedia.org/resource/Wi-fi N/A bathroom http://dbpedia.org/resource/Bathroom http://wordnet-rdf.princeton.edu/wn31/2807731-n Named Entity Dbpedia link WordNet link Madrid http://dbpedia.org/resource/Madrid http://wordnet-rdf.princeton.edu/wn31/9024467-n Alberto Aguilera NH N/A N/A EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 20

  21. Sentiment Analysis extract sentiment words in the context of the identified entities linking to LLOD (SentiWordNet) EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 21

  22. Example: Sentiment Analysis <Author>kekeScotland <Content>We got a good deal compared to other Madrid prices and as we were only using the hotel as a base so we didn't expect too much. Location of the Alberto Aguilera NH hotel was pretty good and the room was quiet and clean. We had a safe and free wi-fi. Only downside was the room was on the small side and the bathroom was tiny. <Date>Sep 10, 2008 <Value>3 <Rooms>4 <No. Reader>-1 <Location>5 <No. Helpful>-1 <Cleanliness>4 <Overall>4 <Check in / front desk>1 <Service>3 <Business service>4 EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 22

  23. Example: Sentiment Words We got a good deal compared to other Madrid prices and as we were only using the hotel as a base so we didn't expect too much. Location of the Alberto Aguilera NH hotel was pretty good and the room was quiet and clean. We had a safe and free wi-fi. Only downside was the room was on the small side and the bathroom was tiny. " Sentiment Word Entity Sentiment Score WordNet link good deal 1 http://wordnet-rdf.princeton.edu/wn31/1123148-a pretty good location 0.75 N/A pretty good Alberto Aguilera NH 0.75 N/A quiet room 0.75 http://wordnet-rdf.princeton.edu/wn31/1922763-a clean room 0.90 http://wordnet-rdf.princeton.edu/wn31/417413-a small room -0.50 http://wordnet-rdf.princeton.edu/wn31/1391351-a tiny bathroom -1 N/A free wi-fi 0.95 http://wordnet-rdf.princeton.edu/wn31/1061489-a safe wi-fi 0.80 http://wordnet-rdf.princeton.edu/wn31/2057829-a EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 23

  24. Lexicon Generator translate entity classes, sentiment words for multilingual analysis enrich with morphosyntactic information (CELEX) convert to lemon and Marl format EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 24

  25. Resulting lexicons (en, de) EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 25

  26. Resulting lexicons (en, de) EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 26

  27. Lexical entries EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 27

  28. Lexical entry: morphosyntactic info EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 28

  29. Lexical entry: DBpedia and WordNet links EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 29

  30. Sentiment polarities using Marl EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 30

  31. Sentiments with context EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 31

  32. German lexicon obtained from translation EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 32

  33. Outline 1. EUROSENTIMENT overview 2. Legacy Language Resources 3. Pipeline for Language Resource Adaptation 4. Tools and Demos EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 33

  34. Available tools and demos Language resources RDF interface: http://www.eurosentiment.eu/dataset Sparql endpoint http://portal.eurosentiment.eu/sparql_demo Resources Navigator http://portal.eurosentiment.eu/lr_navigator_demo EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 34

  35. Sparql demo steps 1. Show the sentiments used with various domain aspects (electronics, en). * domain aspects: CRT_screen, CPU * sentiments: easy, fast, flying * sentiment polarities 2. Show domain aspects (hotel, en) * staff, service, housekeeping 3. Show top sentiment words (hotel, en) * good, great, neat, bully 4. Show positive sentiments (hotel, it) * bello, carino EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 35

  36. Sparql demo steps (contd.) 5. Show negative sentiments (electronics, fr) * difficile, probl me, cher 6. Show links to Dbpedia and WordNet * electronics, en, experience 7. Show translations from BableNet * hotel, en, RO 8. Show translations from the MT approach * hotel, really-confortable, DE * clear-and-readable, ES EUROSENTIMENT use-case Building the Multilingual Web of Data: A Hands-on tutorial 20/10/2014 36

Related


Installations and Commissioned APP Tutorial

Installations and Commissioned APP Tutorial

ivaan
Day 1: Autodesk Inventor Tutorial

Day 1: Autodesk Inventor Tutorial

graciemai
Tutorial for EPDCL Survey App: How to Use Survey Lite App for Data Collection

Tutorial for EPDCL Survey App: How to Use Survey Lite App for Data Collection

satine
Campus Mail Services Package Shipping Tutorial

Campus Mail Services Package Shipping Tutorial

april
AAHS Rutgers Tutorial & Credit by Examination Overview

AAHS Rutgers Tutorial & Credit by Examination Overview

lyra
The Bird Is The Word: A Comprehensive Twitter Tutorial

The Bird Is The Word: A Comprehensive Twitter Tutorial

kaison
New York State Address Confidentiality Program: Application Assistance Provider Tutorial

New York State Address Confidentiality Program: Application Assistance Provider Tutorial

maxim
Audacity Tutorial: Complete Guide on Recording and Editing Audio Tracks

Audacity Tutorial: Complete Guide on Recording and Editing Audio Tracks

leda
Learning Tutorial Observation Methods

Learning Tutorial Observation Methods

frieda
AmeriCorps Member Tutorial on OnCorps Reports 2.0 - Overview and Login Instructions

AmeriCorps Member Tutorial on OnCorps Reports 2.0 - Overview and Login Instructions

nerissa
Excel Tutorial: Evaluating Pay Compression with Compa-Ratios

Excel Tutorial: Evaluating Pay Compression with Compa-Ratios

roselyn
Understanding libfabric: A Comprehensive Tutorial on High-Level and Low-Level Interface Design

Understanding libfabric: A Comprehensive Tutorial on High-Level and Low-Level Interface Design

faris
Interactive Tutorial on Degenerate Perturbation Theory: QuILT Development and Student Difficulties

Interactive Tutorial on Degenerate Perturbation Theory: QuILT Development and Student Difficulties

jac
Micro:bit Robot Entry Tutorial with YahBoom - Buzzer Singing Lesson

Micro:bit Robot Entry Tutorial with YahBoom - Buzzer Singing Lesson

uhtred
Step-by-Step Tutorial on Saving to Your Learner's Drive

Step-by-Step Tutorial on Saving to Your Learner's Drive

ealki
Financial System Journal Entries Tutorial

Financial System Journal Entries Tutorial

ayo_sta
Understanding Active and Passive Voice in Writing

Understanding Active and Passive Voice in Writing

colt_l
Modelling Workflows and Data Domain in System Requirements and User Experience

Modelling Workflows and Data Domain in System Requirements and User Experience

fari_1
SEFA ePledge Tutorial for New York State Employees

SEFA ePledge Tutorial for New York State Employees

yukikogh
Tutorial on Processing Slides and Embedding Animations for JACoW Team Meeting

Tutorial on Processing Slides and Embedding Animations for JACoW Team Meeting

dilo802
Python Review for Data Analytics Tutorial

Python Review for Data Analytics Tutorial

byrd_l
Micro:bit Robot Entry Tutorial - Advanced Lesson on Breathing Lamp with YahBoom (80 characters)

Micro:bit Robot Entry Tutorial - Advanced Lesson on Breathing Lamp with YahBoom (80 characters)

lanc_747

More Related Content