researchmap's Issues
Wrong parameters on function calls
When you deactivate some author/keyword for an award, if you select a new award, those inactives should not be passed to the function, should they?!
investigator-author mismatch
In the grant-pub analysis view, there seems to be a bug in returning the publications. For some reason the matching of names is off.
Investigator is Grbic, but all publications are from Goodale:
Same with another grant from same PI:
With this one the PI is Huner but author is Huang:
Also look at this one. It is from Mike Bauer, but all pubs look like they have to do with chemistry:
Broken view after latest commits on December 31 at around 19:55
It is possibly related to these three commits:
Stop double-click for Non-analyzable awards #7
Read award keywords from arman result instead of couchdb #5
Smaller radious for circles and minimum stoke for lines
As you can see on the terminal on the right hand side, the results is returned, so the analysis.js script is functional.
You can see that even the award information box is not showing any data, which means the UI is the source of issue.
Consistency problem with the list of co-authors
I will work on the issue regarding the long list of co-authors this afternoon and evening.
In the meantime, if Arash has anything to change, or if Paul wants to test and find some bugs, let me know here.
Message box when award is not analyzable, or when _error is 1
I guess it is decided how this should be done now.
Non-analyzable awards
Inactive Publications Keywords Not Showing
Changes to the Analytics control box
I am changing the confidence levels a bit, and hence we might need a modified version of the control box.
What I have in mind is like this:
Assuming we have 3 algorithms to choose, the first look will only show the publication years slidebar on top and then the algorithm selections and nothing more (no aggregation of confidence).
The once the user chooses one of the algorithms, the parameters that are related to that algorithm will show in a dropdown manner. This helps make sure that for the coming algorithm(s), the parameters will not have conflicts, and then we need not deactivate some parameters for those algorithms that do not use those parameters.
So it will be like this:
Analytics
Publications (the slider bar)
Algorithms (Exploration, Keyword Matching, Topic Modelling, etc. each as a radio button)
Then once one selects a radio button, the parameters for that algorithm will show up.
Remove Analysis_test branch
I just removed Analysis_test branch.
keyword matching blank view
publication years not correct in analysis view
password is too easy to hack
This is a very minor issue but I think since we are open to the web now the password should be changed. Currently it's very easy to do a brute force dictionary hack.
submit button stays after switching grants
Package.json jQuery
I assume jQuery should change to jquery.
Removing all authors/keywords
error after deselecting co-authors
double click on a grant. then deselect some co-authors and click submit. then we have only a blank screen. Arash says no data is being returned.
grant circle size and amount
fix filter box in grant-publication view
Parameters not showing change in results
Due to what @pparsons and I discussed a while ago, I am attaching some screen shots after applying some changes.
The first 2 are using linear:
And these 3 are using sigmoid:
However, there is a bug from the db, which when fixed, will show more of this effect.
If @pparsons is in a hurry, I suggest showcasing this functionality using this specific example as it is only returning one publication, and hence, following the changes much easier.
If not, then I am on it after having a late lunch right now/
fix publication circle size
The size of the circles (publications) in the analytics area should be fixed. Currently, the size is set according to the number of publications returned. This makes sense if we want to avoid clutter when many publications are returned (by making the circles smaller). However, the changing size may confuse the user. Additionally, when a small number of publications are returned, the circles are too big and cover the grant circle (see image below). Do you think we should make them a constant size?
Award related information box
Please do not read for the keyword from couchdb.
Instead, use analyzed_award._keyword.
Also, in case you feel it won't take too much space, you could also use analyzed_award._ldaResult for showing extracted topics from the award.
In this case, and for now, only show the first topic, i.e. analyzed_award._ldaResult[0]. This object of the array has tuples in the form of <term,probability>.
It is up to you to decide if we want to show this, and if yes, whether term only, or both.
BTW, for each publication, we can have the same; just in case!
Commit: Increase the gap between grant and publications to fix overlap #1
I am using this updated code, but looks like these gaps are not enough. See attached:
Also, again for the last screenshot, I see no links for most of the publications. Since this is only happening for topic modelling, does it mean that topic modelling is returning a very small radius value? I am not sure if you are putting these values into a scale before assigning them to the like stroke.
publication info for analysis view
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. ๐๐๐
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google โค๏ธ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.