GSOC 2014: 'Label the diagram' application

120 views
Skip to first unread message

Manu Kothari

unread,
Feb 25, 2014, 5:30:11 AM2/25/14
to build...@googlegroups.com
Hello All,
I'm currently doing  my 2nd year engineering in PES University( formerly known as PES Institute of Technology, Bangalore). I'm strongly inclined towards Programming in Android, with an experience of over a year, and find it one of the leading platforms for education. I had a few ideas which can be included in the 'Label the Diagram' application:

1)Integrate voice to text inside the application so that the user can 'Identify the diagram' through speech. I've worked with Speech Recognition in Android through APIs and I think this will be a brilliant user-interactive way of learning.

2) If a student isn't able to label the diagram after 'n' attempts we can also give hints to the students about what part of the body it is, for example, its function, and help the student to learn both at the same time.

3) Instead of only how to label the diagram, sometimes, kids love to colour the diagram. We can give diagrams of the sky, tree leaves, river/sea, etc and give a colour palette containing basic colours. They can drop the appropriate colour on these sheets and make a diagram out of it. Moreover, as some of these colours are obvious , like the sky is blue, we can also teach colours to kids. 

4) User-Interface should be given most importance.

Also, I had a few doubts:

1) What age-group of students are we developing applications for? According to the age , applications will vary.

Thanks!

Rahul Ahuja

unread,
Feb 26, 2014, 2:51:13 AM2/26/14
to build...@googlegroups.com
To answer your question about age group, the app would be for primary school students who study science.


1)Integrate voice to text inside the application so that the user can 'Identify the diagram' through speech. I've worked with Speech Recognition in Android through APIs and I think this will be a brilliant user-interactive way of learning.

2) If a student isn't able to label the diagram after 'n' attempts we can also give hints to the students about what part of the body it is, for example, its function, and help the student to learn both at the same time.

3) Instead of only how to label the diagram, sometimes, kids love to colour the diagram. We can give diagrams of the sky, tree leaves, river/sea, etc and give a colour palette containing basic colours. They can drop the appropriate colour on these sheets and make a diagram out of it. Moreover, as some of these colours are obvious , like the sky is blue, we can also teach colours to kids. 

4) User-Interface should be given most importance.

These are all features, we would like to know your approach towards how you will create the diagram application itself. How would you ask the user to label in a big diagram? How would you validate if the labelling is correct? Please think on these basic questions first.
Message has been deleted

Manu Kothari

unread,
Feb 28, 2014, 5:49:17 AM2/28/14
to build...@googlegroups.com
Hello Rahul,

These are all features, we would like to know your approach towards how you will create the diagram application itself.
 1) The diagram has to be stored as an image in the drawable resources as image. Then we pull the image through drawable:
            private ImageView imgView;                      // use the ImageView class of android
            imgView = (ImageView) findViewById(R.id.imageViewId);         // R.id.imageViewid represents ur image in the drawable folder
            imgView.setImageResource(R.drawable.imageFileId);            // Set the view as the image required.

2)        How would you ask the user to label in a big diagram?

Ok this issue can we tackled in many ways depending upon the needs.

a) We can use tags A,B,C,D on the image and ask the user to input the answer through a drop-drop menu:
For Example, if the question is Label A: 
we can have a drop down menu for the answers.

b) If the user has to drag the word and place it on the diagram only, then we have to pre-determine the co-ordinates 'x' and 'y' on the ImageView of the correct answer. When the user drags the correct tag(which already has the pre-determined x,y stored in it) to the appropriate location only then will the answer be counted.

Again both the above depends on how big the diagram is. If the diagram is bigger, we have to support pinch-in and zoom techniques through the following class: 

3) How would you validate if the labelling is correct?

a) In case (a) above, its very simple, if the user chooses the right option from the drop down menu, its correct.
b) Explained above, if co-ordinates stored match the current co-ordinates ( with +- 5% error ) its correct.

Hopefully, this technique will work. Looking forward about your opinion. 
Thanking You,
Manu Kothari
Reply all
Reply to author
Forward
0 new messages