Can we use splunk for keyword scoring like ELK?

[si_infrastructu]

Hi

We already use Splunk internally and we would like to know if Splunk could be used in scoring uses cases.

We would like to reproduce the same type of example as the one presented in this following article with ELK.

https://www.compose.com/articles/using-query-string-queries-in-elasticsearch/

We have seen that the score command appears with the Splunk Machine Learning toolkit addition but it seems much more complex.

Would it be possible as in the example to have a classified search (in the example with ELK, a scoring on the title of a movie against keywords)

[yuanliu]

Per https://www.compose.com/articles/how-scoring-works-in-elasticsearch/ linked from your reference,

Elasticsearch runs Lucene under the hood so by default it uses Lucene's Practical Scoring Function. This is a similarity model based on Term Frequency (tf) and Inverse Document Frequency (idf) that also uses the Vector Space Model (vsm) for multi-term queries.

Splunk doesn't run Lucene under the hood, but you can implement the model in SPL. (Performance probably won't be the same.)