Automatic Construction of a semantic graph The Sunflower Algorithm - PowerPoint PPT Presentation

Automatic Construction of a semantic graph O. Chabrol, D. Norrestam Project background Entity Disambiguation Automatic Construction of a semantic graph The Sunflower Algorithm Q-numbers EDAN70 Project in Computer Science Implementation Overview Extract information Parsing the dumps O. Chabrol D. Norrestam Merging data Depth Demonstration May,23rd 2016

Automatic Outline Construction of a semantic graph O. Chabrol, D. Norrestam Project background Project Entity Disambiguation background Entity Disambiguation The Sunflower Algorithm The Sunflower Algorithm Q-numbers Q-numbers Implementation Overview Extract information Implementation Parsing the dumps Merging data Overview Depth Extract information Demonstration Parsing the dumps Merging data Depth Demonstration

Automatic Outline Construction of a semantic graph O. Chabrol, D. Norrestam Project background Project Entity Disambiguation background Entity Disambiguation The Sunflower Algorithm The Sunflower Algorithm Q-numbers Q-numbers Implementation Overview Extract information Implementation Parsing the dumps Merging data Overview Depth Extract information Demonstration Parsing the dumps Merging data Depth Demonstration

Automatic Entity Disambiguation Construction of a semantic graph ◮ Basic problem : different entities can have the same O. Chabrol, D. Norrestam name Project background Entity Disambiguation The Sunflower Algorithm Q-numbers Implementation Overview Extract information Parsing the dumps Merging data Depth Demonstration

Automatic Outline Construction of a semantic graph O. Chabrol, D. Norrestam Project background Project Entity Disambiguation background Entity Disambiguation The Sunflower Algorithm The Sunflower Algorithm Q-numbers Q-numbers Implementation Overview Extract information Implementation Parsing the dumps Merging data Overview Depth Extract information Demonstration Parsing the dumps Merging data Depth Demonstration

Automatic Goal of the algorithm Construction of a semantic graph O. Chabrol, D. Norrestam Project ◮ Find concepts linked to every entity background Entity Disambiguation ◮ Using categories linked to article in database The Sunflower Algorithm Q-numbers Implementation Overview Extract information Parsing the dumps Merging data Depth Demonstration

Automatic Construction of a semantic graph O. Chabrol, D. Norrestam Project background Entity Disambiguation ◮ Find the ”best” categories The Sunflower Algorithm Q-numbers Implementation Overview Extract information Parsing the dumps Merging data Depth ◮ We take the most recurrent categories as best categories Demonstration

Automatic Outline Construction of a semantic graph O. Chabrol, D. Norrestam Project background Project Entity Disambiguation background Entity Disambiguation The Sunflower Algorithm The Sunflower Algorithm Q-numbers Q-numbers Implementation Overview Extract information Implementation Parsing the dumps Merging data Overview Depth Extract information Demonstration Parsing the dumps Merging data Depth Demonstration

Automatic Q-numbers Construction of a semantic graph O. Chabrol, D. Norrestam ◮ Allows the link between languages Project background Entity Disambiguation The Sunflower Algorithm Q-numbers Implementation Overview Extract information Parsing the dumps Merging data Depth Demonstration

Automatic Q-numbers Construction of a semantic graph O. Chabrol, D. Norrestam ◮ Unique identifier for entities Project ◮ Every entity (articles and categories) has one universal background Entity Disambiguation Q-number The Sunflower Algorithm Q-numbers Implementation Overview Extract information Parsing the dumps Merging data Depth Demonstration

Automatic Q-numbers examples Construction of a semantic graph O. Chabrol, D. Norrestam Project background Entity Disambiguation The Sunflower Algorithm Q-numbers Implementation Overview Extract information Parsing the dumps Merging data Depth Demonstration

Automatic Outline Construction of a semantic graph O. Chabrol, D. Norrestam Project background Project Entity Disambiguation background Entity Disambiguation The Sunflower Algorithm The Sunflower Algorithm Q-numbers Q-numbers Implementation Overview Extract information Implementation Parsing the dumps Merging data Overview Depth Extract information Demonstration Parsing the dumps Merging data Depth Demonstration

Automatic Overview Construction of a semantic graph O. Chabrol, D. Norrestam Project background Entity Disambiguation The Sunflower Algorithm The algorithm consists of the following steps: Q-numbers 1. Extract information from Wikipedia Implementation Overview 2. For each language Extract information Parsing the dumps Merging data 2.1 Parse information Depth 2.2 Create data structures Demonstration 3. Merge languages and create semantic graph

Automatic Outline Construction of a semantic graph O. Chabrol, D. Norrestam Project background Project Entity Disambiguation background Entity Disambiguation The Sunflower Algorithm The Sunflower Algorithm Q-numbers Q-numbers Implementation Overview Extract information Implementation Parsing the dumps Merging data Overview Depth Extract information Demonstration Parsing the dumps Merging data Depth Demonstration

Automatic Extract information Construction of a semantic graph O. Chabrol, D. Norrestam Project background Entity Disambiguation The Sunflower Algorithm Q-numbers ◮ Have two options Implementation 1. Parse Wikipedia Overview Extract information 2. Use existing dumps Parsing the dumps Merging data ◮ Chose existing dumps Depth Demonstration ◮ DBpedia (2015-10)

Automatic Outline Construction of a semantic graph O. Chabrol, D. Norrestam Project background Project Entity Disambiguation background Entity Disambiguation The Sunflower Algorithm The Sunflower Algorithm Q-numbers Q-numbers Implementation Overview Extract information Implementation Parsing the dumps Merging data Overview Depth Extract information Demonstration Parsing the dumps Merging data Depth Demonstration

Automatic Parsing the dumps Construction of a semantic graph O. Chabrol, D. Norrestam ◮ for every language we have : Project background Entity Disambiguation The Sunflower Algorithm Q-numbers Implementation Overview Extract information Parsing the dumps ◮ and we create : Merging data Depth Demonstration

Automatic Outline Construction of a semantic graph O. Chabrol, D. Norrestam Project background Project Entity Disambiguation background Entity Disambiguation The Sunflower Algorithm The Sunflower Algorithm Q-numbers Q-numbers Implementation Overview Extract information Implementation Parsing the dumps Merging data Overview Depth Extract information Demonstration Parsing the dumps Merging data Depth Demonstration

Automatic Merging data Construction of a semantic graph O. Chabrol, ◮ after merging the three languages: D. Norrestam Project background Entity Disambiguation The Sunflower Algorithm Q-numbers Implementation Overview Extract information Parsing the dumps ◮ we can select the five ”best” categories Merging data Depth Demonstration ◮ with 123 languages

Automatic Outline Construction of a semantic graph O. Chabrol, D. Norrestam Project background Project Entity Disambiguation background Entity Disambiguation The Sunflower Algorithm The Sunflower Algorithm Q-numbers Q-numbers Implementation Overview Extract information Implementation Parsing the dumps Merging data Overview Depth Extract information Demonstration Parsing the dumps Merging data Depth Demonstration

Automatic Depth Construction of a semantic graph O. Chabrol, D. Norrestam ◮ we already have a width concept Project ◮ categories are also part of more general categories background Entity Disambiguation The Sunflower Algorithm Q-numbers Implementation Overview ◮ which leads to depth Extract information Parsing the dumps Merging data Depth Demonstration

Automatic Computation of ratios Construction of a semantic graph O. Chabrol, D. Norrestam Project background Entity Disambiguation The Sunflower Algorithm Q-numbers Implementation Overview Extract information Parsing the dumps Merging data Depth Demonstration

Automatic Demonstration Construction of a semantic graph O. Chabrol, D. Norrestam Project background Entity Disambiguation The Sunflower Algorithm Q-numbers Implementation Overview Interface example Extract information Parsing the dumps Merging data Depth Demonstration

Automatic End Construction of a semantic graph O. Chabrol, D. Norrestam Project background Entity Disambiguation The Sunflower Algorithm Q-numbers Implementation End of the presentation Overview Extract information Parsing the dumps Merging data Thank you ! Any questions ? Depth Demonstration