Creates Knowledge Graph from information processed by "Entity Extraction and Linking", and "Emotion Recognition from Text" module

This MixedEmotions Knowledge Graph was developed by NUIG.

Knowledge Graph provides insight into relations between recognized entities using semantic knowledge from DBpedia. KG module uses entities that are recognised by "Entity Extraction and Linking" module, and extracts relations between the entities from DBpedia. Once the relations are extracted and filtered, they are stored in Elasticsearch database, where using Kibi they are visualized.

This package requires python3.5.

Python libraries:

• elasticsearch
• Flask

DBpedia dumps from http://wiki.dbpedia.org/downloads-2016-04

Minimal list of required files:

• infobox_properties_en.ttl
• instance_types_en.ttl
• persondata_en.ttl

Elasticsearch: 2.4.1

Kibi: kibi-enterprise-standard-4.6.3-2

Type http://localhost:5000/

Kibi already has to contain the source index pattern. In our example trump_tweets.

trump_tweets: index on elasticsearch that contains data processed by Entity Extraction and Linking, [Emotion Recognition from Text] module, and has a field text (that contains original text which was processed).

From Entity Extraction and Linking we use fields:

• entity_linking.URI
• entity_linking.EntityType

From Emotion Recognition from Text we use fields:

• emotions.emotion

http://localhost:5000/configuration PUT

{
"credentials": {
"elasticHost": "localhost",
"elasticPassword": "changeme",
"elasticPort": 9220,
"elasticUsername": "elastic"
},
  "variables": {
"inputIndexName": "trump_tweets",
"inputIndexType": "text_review"
}

}

http://localhost:5000/start GET

http://localhost:5000/status GET

http://localhost:5606/app/kibana#/dashboard/Graph

This development has been partially funded by the European Union through the MixedEmotions Project (project number H2020 655632), as part of the RIA ICT 15 Big data and Open Data Innovation and take-up programme.

http://ec.europa.eu/research/participants/portal/desktop/en/opportunities/index.html