Download presentation
Presentation is loading. Please wait.
Published byJulianna Ludwick Modified over 10 years ago
1
Extraction and Visualisation of Emotion from News Articles Eva Hanser, Paul Mc Kevitt School of Computing & Intelligent Systems Faculty of Computing & Engineering University of Ulster, Magee hanser-e@email.ulster.ac.uk, p.mckevitt@ulster.ac.uk
2
News Visualisation – Emotion Extraction 1 Introduction – What is NewsViz? 2 Background – Related Projects 3 Design & Implementation – The NewsViz Application 4 Prototype Demonstration 6 Testing 7 Relation to Other Work 8 Conclusion & Future Work
3
What is NewsViz? From natural language to visual presentation: NewsViz automatically produces animations from text Input: News Visualisation – Emotion Extraction Online News Article Animation NewsViz System Output:
4
Aim: Animation embedded into news website Objectives: Entertainment Quick overview Emotional aspects >> view website What is NewsViz? News Visualisation – Emotion Extraction
5
The Challenges: 1. Natural Language Processing (computational interpretation of meaning of text) 2. Automatic creation of animations A manageable project: Prototype limited to one topic: football news Focus on determining emotional aspects Reduced to background visualisation What is NewsViz? News Visualisation – Emotion Extraction
6
Syntactic Analysis (based on grammar): Part-of-Speech Tagging (e.g. Qtag) identifying word types such as nouns, adjectives, verbs, … 95-97% correct Qtag Tag-listTagged text Bayern_VB Munich_NP stretched_VBD their_DPS lead_NN at_PRP the_AT top_NN as_CJS Hamburg_NP suffered_VBD a_AT tragic _JJ surprise_NN home_NN loss_NN._. PRPpreposition JJadjective, general NNnoun, common singular NNSnoun, common plural NPnoun, proper singular VBverb, base from VBDverb, past tense... http://www.english.bham.ac.uk/staff/omason/software/qtag.html Related Projects News Visualisation – Emotion Extraction
7
WordsEye: Creates static 3D scenes from text input http://www.wordseye.com WordsEye Related Projects News Visualisation – Emotion Extraction
8
WordsEye – Description and Rendered Image http://www.wordseye.com The skiff is on the ocean. The grassy mountain is 20 feet behind the boat. The dog is in the boat. The fishing pole is two feet in front of the dog. The bottom of the palm tree is below the bottom of the mountain. It is 20 feet behind the boat. WordsEye Related Projects News Visualisation – Emotion Extraction
9
More Syntax Analysis: Structure of Sentences Dependency Parser (e.g. used in WordsEye) Finding relations between words and phrases Dependency rules Who? Does? What? http://www.wordseye.com WordsEye Related Projects News Visualisation – Emotion Extraction
10
http://www.wordseye.com WordsEye Related Projects News Visualisation – Emotion Extraction Graphical Database in WordsEye 3D objects, their attributes (colour, size, surface)
11
Semantic Analysis (based on meaning): Lexical Knowledgebase (e.g. WordNet) sets of synonymous words and basic semantic relations Semantic Relation Synonymy (similar) Antonymy (opposite) Hyponymy (subordinate) Meronymy (part) Troponomy (manner) Entailment Examples pipe, tube sad, unhappy wet, dry rapidly, slowly maple, tree tree, plant wheel, car whisper, speak divorce, marry Syntactic Category N, V, Aj, Av Aj, Av, (N, V) N V http://wordnet.princeton.edu/ WordNet Related Projects News Visualisation – Emotion Extraction
12
The Story Picturing Engine: matching keywords + image regions step 1: filtering out common words (a, the, of, …) step 2: identification of proper words (places and people involved) step 3: similarity count of remaining keywords (words with too many meanings are too vague) … further steps for image processing Story Picturing Engine Related Projects News Visualisation – Emotion Extraction
13
Example text on walk through Paris H = highest ranked images, L = Lowest ranked images Story Picturing Engine Related Projects News Visualisation – Emotion Extraction
14
NewsViz Architecture NewsViz News Visualisation – Emotion Extraction
15
Emotion Visualiser NewsViz News Visualisation – Emotion Extraction
16
Graphics Database Abstract Visuals for 4 Emotions 2 - boring 4 - happy 3 - tense 1 - sad NewsViz News Visualisation – Emotion Extraction
17
Word Lexicon with Emotion Indices challenges 3 <!– tension 3 … home 4 <!– happy 1 gaps 1 <!– sad 2 NewsViz News Visualisation – Emotion Extraction
18
Summarization Options NewsViz News Visualisation – Emotion Extraction
19
Demonstration News Visualisation – Emotion Extraction
20
Demonstration News Visualisation – Emotion Extraction
21
Demonstration News Visualisation – Emotion Extraction
22
Demonstration News Visualisation – Emotion Extraction
23
Demonstration News Visualisation – Emotion Extraction
24
Demonstration News Visualisation – Emotion Extraction
25
Demonstration News Visualisation – Emotion Extraction
26
Demonstration News Visualisation – Emotion Extraction
27
Demonstration News Visualisation – Emotion Extraction
28
Procedure NewViz performance evaluated against human interpretation: 1. General mood course (3-5 emotions per text) 2. 1-2 Emotions per sentence types of emotion extraction error Falsely detected emotion : 0 points Missing emotion : points depending on significance Overall feeling represented, 2-3 points Similar emotion : 4 points Exact emotion: 5 points Testing News Visualisation – Emotion Extraction
29
Results Course of moods mostly identified correctly Word-by-Word method highest correctness but too fine grained for animation Best results with both (adjective and nouns) Testing News Visualisation – Emotion Extraction Method Word bySentenceThresholdaverage Wordbased23 Word type correctgraincorrectgraincorrectgraincorrectgrain adjectives3.125123.257.52.37551.252.32.5 nouns3.875312.6259.32.8751424.82.844 both4332.759.53.5181.5102.938 average3.667252.8758.82.917121.5835.7
30
Summary Emotional interpretation of online news articles Course of moods depicted in abstract 2D animations Different methods of language processing Satisfactory outcome User Evaluation Appreciation of animations as quick overviews Future Work Extension of knowledge bases Inclusion of different topics Improvement of summarisation, e.g dependency parser Conclusion & Future Work News Visualisation – Emotion Extraction
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.