Wiki Mind Mapping

20
Wiki Mind Mapping Wiki Mind Mapping Harshit Mittal (IIT-B) [email protected] Aditya Tiwari (IIT-B) [email protected] Akhil Bhiwal (VIT University) [email protected] 1

Transcript of Wiki Mind Mapping

Page 1: Wiki Mind Mapping

Wiki Mind MappingWiki Mind Mapping

Harshit Mittal (IIT-B)[email protected]

Aditya Tiwari (IIT-B)[email protected]

Akhil Bhiwal (VIT University)[email protected]

1

Page 2: Wiki Mind Mapping

Project IdeaProject Idea

Represent the textual information in graphical form which is easier to understand and more intuitive to read. The visual representation should be able to summarize the text.

2

Page 3: Wiki Mind Mapping

Research GoalResearch Goal

Use of phrases to represent semantic information.

Hierarchical representation of information of a given text

3

Page 4: Wiki Mind Mapping

Mind mapsMind mapsA mind map is a diagram used to

represent words, ideas, tasks, or other items linked to and arranged around a central key word or idea.

Example Mind map in the next slide.

4http://en.wikipedia.org/wiki/Mind_maps

Page 5: Wiki Mind Mapping

Mind mapMind map

5http://www.spicynodes.org/blog/2010/05/21/stuff-we-like-climate-change-mind-maps/

Page 6: Wiki Mind Mapping

WhatWhat’s the difficult part?’s the difficult part?

We can’t represent information from any article in mind-map as it is. That would make it incoherent and clumsy.

Phrase extraction

General rules of grammar don’t apply here.

6

Page 7: Wiki Mind Mapping

Possible SolutionPossible Solution

Develop new linguistic rules for representation of text in visual form.

Use existing summarization tools to generate summary and try to represent that in mind-map.

7

Page 8: Wiki Mind Mapping

How we did it.How we did it.Pulling out the article section wise from the

Wikipedia page.

Parsing each section sentence wise using the Stanford parser.

Extracting “relevant” phrases using Tregex (another Stanford tool).

Putting these phrases into a mind map, section wise.

8http://nlp.stanford.edu/software/tregex.shtml

Page 9: Wiki Mind Mapping

Extraction of relevant informationExtraction of relevant informationIdentifying subtrees from the parse tree of a

sentence that are important.

This was done using a few heuristics like: ◦ Presence of a superlative adjective in a noun phrase

9http://nlp.stanford.edu/software/tregex.shtml

Page 10: Wiki Mind Mapping

Extraction of relevant informationExtraction of relevant informationPresence of a cardinal number in a noun

phrase

10http://nlp.stanford.edu/software/tregex.shtml

Page 11: Wiki Mind Mapping

Extraction of relevant informationExtraction of relevant information

Matching of a particular verb to the bag of verbs that were considered relevant for a particular article. For example : for the history section, verbs like find , discover, settle, decline were considered “more useful”, as compared to words like derive, deduce etc. which were considered useful for some other section.

11

Page 12: Wiki Mind Mapping

Ex : The name India is derived from Indus.

12

Extraction of relevant informationExtraction of relevant information

http://nlp.stanford.edu/software/tregex.shtml

Page 13: Wiki Mind Mapping

13

Code Generated Mind MapCode Generated Mind Map

Page 14: Wiki Mind Mapping

EvaluationEvaluation

14http://en.wikipedia.org/wiki/Precision_and_recall

Page 15: Wiki Mind Mapping

EvaluationEvaluationSurvey based:

Asking a person to generate 10 questions from given article.

Asking another person to answer those question with the help of mind-map.

Repeating the same exercise in reverse manner for another article.

15

Page 16: Wiki Mind Mapping

ObservationsObservationsPros:◦ Extraction of right information with high

accuracy.

◦ Concept of phrase extraction works well.

◦ High precision value were obtained (between 0.5-0.75).

16

Page 17: Wiki Mind Mapping

ObservationsObservationsCons◦ Information presented in mindmap of low depth

is clumsy.

◦ Low recall value (0.2 – 0.4)

◦ Linking of node phrases with their apt description.

◦ Heuristics defining “important phrases” need to be refined.

17

Page 18: Wiki Mind Mapping

LimitationsLimitationsBag of words and Tregex expressions is

hand-coded instead of machine learned.

Garbage phrases are being generated.

Level of hierarchy is limited to 3.

18

Page 19: Wiki Mind Mapping

Future workFuture workUsing machine learning to determine the

important keywords for a given sentence.

We want to explore the possibility of finding patterns in subtree expressions using machine learned approach.

Refinement of generated phrases.

19

Page 20: Wiki Mind Mapping

ReferencesReferences

http://en.wikipedia.org/wiki/Mind_mapshttp://en.wikipedia.org/wiki/Precision_and_recallTool : Stanford Parser and Stanford Tregex Match

http://nlp.stanford.edu/software/tregex.shtml

20