Created for Elasticsearch X Code for Cause Hackathon. (Winner for the hackthon - devpost project)
This is the backend for my MemeTastic Application.
-
Scrapes top 100 memes from subreddit
memesusing reddit's public API. -
Adds the
linkand thetitleof those 100 memes inelasticsearchwhich can then be visualised usingkibana. -
Elasticsearchhas been hosted onGoogle Cloud. -
The first two steps are repeated after every 30 minutes using a CI/CD pipeline created using
GitHub Actionsto make sure that the Application stays up-to-date.
- This also uses ngrams, implemetation is given below (implemeted using
kibana)
PUT dankmemes
{
"settings": {
"analysis": {
"analyzer": {
"autocomplete": {
"tokenizer": "autocomplete",
"filter": [
"lowercase"
]
},
"autocomplete_search": {
"tokenizer": "lowercase"
}
},
"tokenizer": {
"autocomplete": {
"type": "edge_ngram",
"min_gram": 1,
"max_gram": 10,
"token_chars": [
"letter"
]
}
}
}
},
"mappings": {
"properties": {
"title": {
"type": "text",
"analyzer": "autocomplete",
"search_analyzer": "autocomplete_search"
}
}
}
}