Susi AI Server

MentorHong Phuc Dang (hpdang)
Project Websitehttp://asksusi.com
Project Repositoryhttps://github.com/fossasia/susi_server/
Suitable for Beginners?no
TagsAI machine learning rules based AI personal assitant
Stateaccepted
Applications (1st Choice)3 (3 submitted | 0 in-progress)
Applications (2nd Choice)4 (3 submitted | 1 in-progress)
Code of Conducthttp://coc.fossasia.org
LicenseGNU Library or "Lesser" General Public License version 2.1 (LGPL-2.1)

Project Description

Susi AI

Susi AI is a server application which is able to interact with humans as a personal assistant. Susi AI is using a wide variety of sources and data to provide answers or perform actions demanded by the user. Susi AI is not only able to answer questions, but also to ask for additional information in order to perform the desired outcome.

An automatic deployment from the development branch at GitHub is available for tests here https://susi-server.herokuapp.com

Communication

Please join our mailing list to discuss questions regarding the project: https://groups.google.com/forum/#!forum/loklak

Our chat channel is on gitter here: https://gitter.im/fossasia/susi_server

Susi AI also has an Android client and a Web App client using Angular JS.

The project is run by the FOSSASIA team.

Project's Requirements

Technologies Used

We are looking for students with good knowledge in the following technologies:
* Java
* HTML
* Javascript
* CSS
* Bootstrap
* Elastic Search

Other Skills

Participants in the program are expected to:
* Be communicative and follow up on our channels on open questions
* Learn about Best Practices regarding merges, pull requests, issues and help to constantly improve the development flow

Community

We would also like to see participants helping other newcomers in the project.

Participation in FOSSASIA Summit

Successful participants are encouraged to apply for the FOSSASIA Summit in Singapore in March 2018.

Tasks And Features

Enhance the Console Rule Learning App for Susi Server

A basic console learning app is implemented in the server already (see here). The goal of this project is to make it easy now to create new console rules a) to connect new APIs to Susi and b) to add the rules itself.

The rules are the method to attach external knowledge to Susi. To check and store new console rules, there is now the rule learning backend api, the details are described here: https://github.com/fossasia/susi_server/blob/development/src/org/loklak/api/learning/ConsoleLearning.java#L45-L163

To store a new console rule, the user has to enter at least the following entries:
- the external json API address
- an example input for that API
- a name for the new console rule

Furthermore, the user must submit a jsonpath which points into an result array within the JSON from the console API. The new back-end process supports the user by finding that jsonpath automatically.

What we need:
- an input window for those three options
- a 'verify' button which calls the backend. That will return a lot of information together with the jsonpath which the user can edit.
- a 'store' button which can be used when the user is satisfied with the back-end verification.

You can find the corresponding issue for this project here: https://github.com/fossasia/susi_server/issues/73

Requirements

Please have a look on our website at http://fossasia.org/apply.html to learn about how we run coding programs. During the Summer of Code we would like students to follow the outline requirements of the RGSoc program as well as requirements of the developer team that enable us to collaborate.

These include:
* Participate full-time from July 1st to September 30th 2017
* Continuously work on the Open Event project.
* Keep track of your work with a short, daily summary in form as a scrum to the mailing list
* Keep regular contact with your mentor and the development team and abide to reaction times for communication.
* Provide a weekly blog entry with images and links explaining/sharing your work
* Provide an email address and Twitter account for social and promotional events.
* Participate and organize non-coding related community events, e.g. a meetup in the area where you live
* Agree to have published team information on our website and blog (e.g. team introduction, team blog posts). This includes some form of visual material of yourself.

Comments

You must be logged in to comment on this project.


Soumya, Thursday, February 23, 19:46 UTC

Hello,
We looked through the project and loved the idea of developing our own Allo. We are BTech students from New Delhi, India. We'd love to contribute to this project and expand our knowledge in ML, AI and NLP.
I've already joined the Gitter channel and started clearing my concepts about the project.


Mario Behling, Saturday, February 11, 13:37 UTC

Hi, please join our Gitter channel https://gitter.im/fossasia/susi_server and start to learn about the project on Gitter. What is your Github?


Laura RGSoC, Friday, February 3, 11:34 UTC

This would be a good project for RGSoC, happy to accept it. \o/


Y S Ramya, Friday, February 3, 11:19 UTC

Hi, we are BTech students studying in India. We are very interested in the field of Machine learning and Artificial Intelligence and would love to learn more about it. It is a request to the RGSoC team to approve this project as we are very would like to make it, hands down, our first preference.


Akarsha Sehwag, Friday, February 3, 09:44 UTC

We are really willing to work on this project.!! Is there anything we can do to get it accepted.?