Tool developed during Houman Hadian's doctoral project as code base for social APIs ETL and analytics. @UAlg.pt
Socialytics, is intended to provide batch and realtime (LAMBDA) ETL functionality on Instagram service APIs (private web API) in order to analyse brand engagement on Instagram via analytical models presented on Houman Hadian's doctoral project research papers. This is an open source contribution in hope of producing side common benefits along with academic & scientific values. Soon to be published separately as NPM package with a more techy name indeed! We beleive that respecting diversity of tech tools in the wild there are still few full functional FOSS tools for browsing and analytics on social networks out there. Seeing a gap that could be filled makes the purpose to make a tool that can be reused for research, academic and scientific purposes. Noticible impacts on branding (place, product & service), tourism, marketing and behaivioral analytics sciences are envisioned and sought.
Notice: Work in progress!
The front end application is primitive HTML just for sake of testing and
needs to be completed. For now just links to API endpoints are available.
and you have to change the parameters directly inside URI. The login process is still shakey and PM2 integration is being completed or even get removed because of the extent of resources and concerns ratio.
TODO:
1- Add Instagram login form as start page for frontend and store the session-id and csrf_token on server side (with security in mind!)
2- Add UI components for parameters
3- Implement database loading (NeDB for time being with support for MongoDB, Postgress, ElasticSearch and Redis).
4- Add progress , start and stop for every ETL operation.
5- Add export to CSV/JSON option to every ETL process.
6- Add Analytics dashboards and components.
7- Add analytical models and algorithms.
//---------------------------------------------------
Quick start:
1- Clone this project
2- Optionally-- Run: npm install -g pm2
3- Config your session_id from your browser on Instagram website (within Instagram cookies!).
4- In CMD go to project root folder
5- Run: npm install then Run:npm start (or optionally Run: pm2 start src/)
6- Go to localhost:8080 and start analysing Instagram!
Contributors:
M.Ghiasi (emg110@gmail.com), H.Hadian (a58976@ualg.pt)
Special thanks to these great open source projects which paved the client's starting point but since changes required to make the tool complete , comprehensive and calibrated for analytics hence they have undergone great changes which left no place for branching and backward contribution via pull requests thus here we announce our gratitudes and in turn give back the whole effort to community, free and open source: