Refine Hit Highlighting in azure-db

[k-allagbe]

Our Azure Cognitive Search is currently highlighting frequent but less meaningful words (such as basic conjunctions or articles), which can reduce the relevance and clarity of search results. The aim is to adjust the hit highlighting feature to emphasize more significant and contextually relevant terms.

Tasks:

• [ ] Investigate and apply query adjustments to prevent highlighting of frequent but less relevant words using the highlight parameter.
• [ ] Conduct thorough testing to ensure the modifications enhance the relevance and clarity of highlighted search terms.

Acceptance Criteria:

• The updated hit highlighting omits frequent but less significant words, focusing instead on contextually relevant terms.
• The search experience is enhanced, with highlights providing clearer and more meaningful insights into search results.
• All changes are thoroughly tested, confirming their effectiveness in refining the search results presentation.

Suggestions

• Consider the creation of a custom analyzer in the Azure Search index to systematically exclude common stop words.
• Modify search queries to selectively apply hit highlighting to specific fields, reducing focus on less relevant content.
• Tweak the configuration of highlight fragments to more effectively capture and present key portions of text.
• Customize highlight tags to improve the visibility and distinctiveness of search result highlights.