Lemmatization in Neo4j within cypher query

[sudarshan_kotha]

I want to find out lemma of the given word using Cypher query and use the result to perform next level of task. I've found out that Neo4j has

1."Full text indexes searches" - https://neo4j.com/docs/cypher-manual/current/administration/indexes-for-full-text-search/
2. Stanford NLP - https://github.com/graphaware/neo4j-nlp-stanfordnlp

But both of them creates lots of NLP related datapoints such as POS, NER. I don't want to store any information but use them in cypher query. Can you please help me on how it could be achieved? Eg: "credits" -> "credit"

Neo4j Desktop Graph version: 3.5.18
APOC LIbrary: 3.5.0.12

Thanks in advance!

[jggomez]

sudarshan.kothari:

https://neo4j.com/docs/cypher-manual/current/administration/indexes-for-full-text-search/

Hi, I don´t understand your question...

You could take different words and get the lemma with spacy. You can do an APOC and it receives a parameter, the word and the result will be the lemma.

Lemmatizer

Lemmatizer · spaCy API Documentation

Assign the base forms of words

[sudarshan_kotha]

Just to check if my understanding is correct - You are asking me create a model kinda object and perform lookup everytime, I need a lemma of the keyword from the model?

To paraphrase my question:
I'm making a dynamic cypher query, which in-takes a set of words (generally for 5 words) and finds out their lemma of individual words and which is required for further processing.
For example: "Best offers for vacation this summer"
Output interested: ['Best', 'offer', 'for', 'vacation', 'this', 'summer']

Let me know, if this is still not clear.

Thanks for your time!