Twitter Mining in Python and Web UI
- or -
Post a project like this£200(approx. $250)
- Posted:
- Proposals: 6
- Remote
- #1868628
- OPPORTUNITY
- Expired
Customized Mobile apps | Website Design & Development | Digital Marketing |E-commerce Solutions
Ahmedabad
160160916124311886115194622620253682061854
Description
Experience Level: Intermediate
Estimated project duration: less than 1 week
General information for the business: Twitter Text Mining
Kind of development: New program from scratch
Description of requirements/functionality:
A program for mining twitter for individual and multiple keywords and mentions using the twitter API and storing the JSON output. Due to the limit imposed by twitter on the API the program will have to rotate around multiple API keys (four?), I require the program to be able to collect a relatively large number of tweets and store them in an appropriate database. I intend to mine several hundred thousand tweets (at least) and their associated metadata; tweet, tweet ID, username, location, geo location, language, date created.
What is this for? I require the program to be able to mine and store tweets relating to different individual and combinations of keywords. For example: BMW, new BMW, broken down BMW, etc. In this case we are collecting neural BMW tweets, positive BMW tweets and negative BMW tweets.
The ultimate aim is to the create a web UI pulling from the database of tweets to display individual tweets and provide the website user with four options to label the tweet; positive, neural, negative and irrelevant. The point is to "crowd source" the hand annotation of the tweets (it has to be by hand). The web UI will simply be a single page website with those four buttons and the tweet displayed.
The initial program will be run from my computer to collect the tweets, the web UI is to run on a website hosted on a virtual machine by me at a later date so colleagues and others can help annotate the tweets. The web UI requires logic to label a tweet as "confirmed" or "agreed" once three separate users agree on its sentiment (positive, neural, negative or irrelevant), after this it is no longer shown to users. The web interface is very simple, literally just the tweet and those four buttons, designed as user friendly as possible - encouraging the user to annotate as many as possible by making it easy. Bigger buttons, easy to read text, visually appealing flat design.
I would prefer the initial program to be written in python, the other elements can be in any appropriate language. The choice of database is at your discretion.
I encourage the use of any frameworks, boilerplate, libraries or open source code. Please follow best practices and standards for the languages and tools you're using - including database creation. Please comment the code appropriately so I can tinker more easily at a later date. When using libraries and frameworks, please to popular and up-to-date ones. Please don't use obsolete frameworks.
Please note I only require what is described here. Two elements: the program to mine the tweets and the web element for web users to categorise them by hand. The key is that this will be done by hand eventually. I am building a database to train a neural network, it cannot be done by any other neural network or machine learning library. It needs to be 95-99% accurate.
You are creating artefacts which upon final payment transfer all rights to the recipient without limitation or license.
Please don't submit wildly inflated bids... I am fully aware of the skills this project requires. Feel free to increase the price dependant upon your experience, I appreciate skill and experience. However I will not consider seemingly randomly high bids. You could easily follow a tutorial for the twitter mining element, and basic programming knowledge is required for the database and web UI. I am looking for someone with intermediate experience, following best practises etc, hence the reasonable budget of around £200.
If you have any questions please ask.
Thanks for your time.
Kind of development: New program from scratch
Description of requirements/functionality:
A program for mining twitter for individual and multiple keywords and mentions using the twitter API and storing the JSON output. Due to the limit imposed by twitter on the API the program will have to rotate around multiple API keys (four?), I require the program to be able to collect a relatively large number of tweets and store them in an appropriate database. I intend to mine several hundred thousand tweets (at least) and their associated metadata; tweet, tweet ID, username, location, geo location, language, date created.
What is this for? I require the program to be able to mine and store tweets relating to different individual and combinations of keywords. For example: BMW, new BMW, broken down BMW, etc. In this case we are collecting neural BMW tweets, positive BMW tweets and negative BMW tweets.
The ultimate aim is to the create a web UI pulling from the database of tweets to display individual tweets and provide the website user with four options to label the tweet; positive, neural, negative and irrelevant. The point is to "crowd source" the hand annotation of the tweets (it has to be by hand). The web UI will simply be a single page website with those four buttons and the tweet displayed.
The initial program will be run from my computer to collect the tweets, the web UI is to run on a website hosted on a virtual machine by me at a later date so colleagues and others can help annotate the tweets. The web UI requires logic to label a tweet as "confirmed" or "agreed" once three separate users agree on its sentiment (positive, neural, negative or irrelevant), after this it is no longer shown to users. The web interface is very simple, literally just the tweet and those four buttons, designed as user friendly as possible - encouraging the user to annotate as many as possible by making it easy. Bigger buttons, easy to read text, visually appealing flat design.
I would prefer the initial program to be written in python, the other elements can be in any appropriate language. The choice of database is at your discretion.
I encourage the use of any frameworks, boilerplate, libraries or open source code. Please follow best practices and standards for the languages and tools you're using - including database creation. Please comment the code appropriately so I can tinker more easily at a later date. When using libraries and frameworks, please to popular and up-to-date ones. Please don't use obsolete frameworks.
Please note I only require what is described here. Two elements: the program to mine the tweets and the web element for web users to categorise them by hand. The key is that this will be done by hand eventually. I am building a database to train a neural network, it cannot be done by any other neural network or machine learning library. It needs to be 95-99% accurate.
You are creating artefacts which upon final payment transfer all rights to the recipient without limitation or license.
Please don't submit wildly inflated bids... I am fully aware of the skills this project requires. Feel free to increase the price dependant upon your experience, I appreciate skill and experience. However I will not consider seemingly randomly high bids. You could easily follow a tutorial for the twitter mining element, and basic programming knowledge is required for the database and web UI. I am looking for someone with intermediate experience, following best practises etc, hence the reasonable budget of around £200.
If you have any questions please ask.
Thanks for your time.
Kristofor B.
93% (3)Projects Completed
4
Freelancers worked with
4
Projects awarded
50%
Last project
6 Mar 2018
United Kingdom
New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-
There are no clarification messages.
We collect cookies to enable the proper functioning and security of our website, and to enhance your experience. By clicking on 'Accept All Cookies', you consent to the use of these cookies. You can change your 'Cookies Settings' at any time. For more information, please read ourCookie Policy
Cookie Settings
Accept All Cookies