Determination of main belonging cluster for coocurences Keywords

CorText Manager Q&A forumCategory: Network mappingDetermination of main belonging cluster for coocurences Keywords
Beadavi asked 4 years ago

I made a coocurence keyword map on WOS references (KW author and ISI keywords) and asked that each reference be assigned to a unique cluster. How is the main belonging cluster determined, knowing that a reference can belong to several clusters?
Thank you in advance

Lionel Staff replied 4 years ago

See below

1 Answers
Lionel Staff answered 4 years ago

Dear Beadavi,
As an example, I will take a corpus of documents with textual information (e.g. titles), where I run a lexical extraction. This lexical extraction extracts relevant keywords for each document.

Running a Mapping Network script with Extracted Keywords – Extracted Keywords, will build a semantic landscape with clusters of keywords. The documents will be projected onto those clusters through their extracted keywords.

So, the “Assign a unique cluster to each record (best match)” is at the document level It makes CorText Manager able to choose the best clusters (hightest weight) for each document.
This weight (relation between a document and a cluster) is calculated by summing the tf-idf of each extracted keyword for a cluster pondered by a measure of the importance of the keyword in that cluster.

I hope it helps!

Beadavi replied 4 years ago

Hello ,
Thank you for your response.
So the attribution of a reference to a cluster on the basis of the author keywords and ISI keywords is relevant, even if the number of occurrences of the terms in the document is always 0 or 1 since it is not free text?
Thank you in advance