Skip to content

Content Search Engine

Jump to bottom
Jamarri Green edited this page Dec 8, 2023 · 1 revision

**Content Search Engine **

Content Search Engine

Preexisting Work

  • SearchByQuestion: Briefly speaking, it takes in a user-inputted string (that doesn't necessarily have to be a question, given that during a query, an assumption is simply made that the string is such), and checks against what's currently in ACMAS' database (viewable in Docker) to see if there are any .txt files that contain the "question" passed. Granted, the "question" must exist in verbatim (i.e. there can be no partial matches, which is a strong disadvantage, and why SearchByKeyword exists).

  • SearchByCourse: Works similarly to the above in that it searches for files with similar parameters, with the difference being that a user must instead pass a university, course, and assignment type (i.e. exam, practice, etc.). Subsequently, what this means for any found results is that the nature of said results are verifed via additional matching parameters to the query (the university/course/assignmentt in this case) compared to the above algo, and, also unlike the above algo, is a bit more specific regarding how it handles uploaded files.

New Algorithm

  • SearchByKeyword: Removes SearchByQuestion's drawback of requiring a verbatim "question" matching the query to exist in the database. However, there are a lot of considerations to be made when using this algo in particular (link in --considerations-- section).

Considerations

Each one of these links will take you to a different file whose contents entail various different dilemmas that you have to take into consideration when approaching the titular topics (such as dilemmas that arise when trying to cache peevious search results):

Clone this wiki locally