pshahid / customer_hunter Goto Github PK
View Code? Open in Web Editor NEWUse Social media analytics to find and convert dissatisfied customers
Use Social media analytics to find and convert dissatisfied customers
When you first load the page in production sometimes the Tweet page is null. But it works again when you restart the core process.
We need a UI and back-end portion for showing users what notifications they have, like new follows, favorites, etc.
SQLAlchemy is a much better ORM for us, and it doesn't have some of the gotchas of Peewee. It's also better for production. But, the learning curve is steep so I gotta wait to do this, can't do it right now.
A user must be able to login to any number of their Twitter accounts through our application and authorize it to do actions on their behalf.
Ability to change account information like username, password, email from when you're logged in to ML Systems.
Currently I'm pretty sure we don't store lat, longs in the data base. We need to get on that.
Boomer mentioned that it'd be useful to have the coords of users
Users on the client should be able to reply to a twitter status via an input box. It should be similar to the twitter one, too, where it auto-formats tags, hashes and links.
We want the experience of this app similar to the twitter experience where things are auto-formatted when they're put in
Right now we're blocking whenever we call stream.next(). This is a problem, because it will block Twisted's eventloop from running. Instead, we should make the consume method asynchronous in the concrete strategy implementations
For the consumer-facing website we need:
Sometimes we see a link being left in a status. I observed 2 instances of this in the wild, but it's hard to find. The conditions appear to be: when it's a retweet/reply, when there's 2 links, 1 is media and the other is an actual link.
Not really sure how to fix until I have more info.
We need to use site and user streams to stream in data for any particular connected client. Unfortunately, this is a lot more work on the back-end. The reason why it's necessary is to avoid being rate-limited by Twitter and also allowing us to stream data to users in real-time for their timelines instead of polling.
Similar to Tweetdeck we're going to add a column for saved searches and timelines for their logged in accounts. This will mimic the UI feel for Tweetdeck so it won't feel so foreign
Angular-ui-router is fine, but we really don't need it and it was a mistake to include it in the first place. It ended up becoming an anti-pattern and difficult to remove since it was so ingrained. It needs to be removed in favor of ngRoute
The back-end needs to resume classifying tweets. At this point we need to send tweets to clients if they're classified '1' in the database.
On appendixg.com server, using Nginx to serve static Test.html, something happens to the browser-side of the connection where it stops receiving events from the server.
I'm not sure as to why this is happening, it could be any of:
This needs to be investigated and triaged so that it doesn't degrade user experience. As a workaround, consider using pings from the client side to give a heartbeat to the TCP connection such that it won't close after a given time.
Need to test this on localhost as well to see if it's just an nginx problem or a websocket problem.
Because that would be nice.
Right now we're using Flask's development server, which is not the way to run in production. It's fine when we do it locally, but not when deployed.
Expand the tweet table columns to include nearly every field captured by twitter. A lot of this metadata is useful to mine later on for analytics and it'd be better to store it.
Need to get our mongo query back in reverse-chronological order, and be able to use a cursor to cycle through data via infinite scroll on the client.
For the Python back-end we need to incorporate the Flask webapp framework.
Move it into its own python process. It's much easier to manage if we have multiple python processes doing their own things. Right now we have one which is the Flask webapp that serves the website and one that does both consuming and websocket serving. The websocket server needs to be its own thing.
We need to train the restaurant data set that we've been amassing. Accuracy is of paramount importance so it's good to start early on this. Don't rush it either.
We need to add a workflow for when you forget your password.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.