search

alex9311 / information-retrieval

TU Delft, Masters Software Technology, Information Retrieval, 3rd Quarter 2015
1 stars 2 forks source link

Scope discussion #39

Closed HDking closed 9 years ago

HDking commented 9 years ago

Scope

Need vs. Nice

All done by Monday!

Need

millenniumproof commented 9 years ago

You make it look like I need to test my sanity! My brain is perfectly fine, thank you :P

HDking commented 9 years ago

haha sorry!

alex9311 commented 9 years ago

How is everyone doing on this stuff?

millenniumproof commented 9 years ago

I have the code for mysql to csv for crowdflower and csv to mysql for the sanitation check part. Haven't tested it on the server yet though.

There is also a way to dynamically add data to a job on CrowdFlower and receive the results using JSON. I want to try and get that to work but I might not have the time today.

PetervB commented 9 years ago

I have put quite some time in lucene this weekend, I got to the part where I am using the MoreLikeThis method of lucene that is used to find similar documents, however this is currently only finding directly the same documents and I don't know why...

HDking commented 9 years ago

I've tried to learn it all myself, but I need some help from someone who knows some stuff about PHP and SQL stuff to get the voting page work.

2015-03-15 16:17 GMT+01:00 PetervB notifications@github.com:

I have put quite some time in lucene this weekend, I got to the part where I am using the MoreLikeThis method of lucene that is used to find similar documents, however this is currently only finding directly the same documents and I don't know why...

— Reply to this email directly or view it on GitHub https://github.com/alex9311/TUD-Information-Retrieval-Group-02/issues/39#issuecomment-81085697 .

Hao Dinh Korvezeestraat 442 2628 DV, Delft +31651272204

millenniumproof commented 9 years ago

Whoops. Committed to the Master. The Crowdflower stuff is in separate files anyway, so it's okay (^-^). I couldn't get it to work dynamically. I tried out stuff with cURL requests but I couldn't get a response. I can still look into it some more but I think the Similarity check (and even the FaceBook login) have higher priority because the user is not involved in the CrowdFlower part and it works the way it is now (just a bit of a pain in the ass to do the uploading and downloading manually).