Detection of offensive language in oline social networks

Hello, I'll need few more details about project. Which Social Media are you trying to parse/scrap for offensive words? Twitter offers API and it will be easier to achieve goal using API then by scrapping it. I'm also willing to send working demo and if it works as it should you can award me for the project. Thanks, Stevan

Hi. I'm a programmer from Russia with strong skills in Python. You can see the full list of my skills on my profile page. My plan is following. We can include probabilistic approach that will allow to detect not only offensive words, but their derivatives also. For example, basic comparison won't detect offensive word in phrase "go afucka yourself". But if algorithm count a simple probability as distance from one word to another, it will be able to mark the word in the middle as suspicious at least. That's them main idea. If you're interested, feel free to contact me. All the best, Sergey.

Your project sounds very interesting. I have a master in Machine Intelligence from Aalborg Unviersity, Denmark, which is a field very focused on training classifiers for different machine learning tasks. There is not yet enough details for me to make a very accurate estimate of the task at hand, but I would love to discuss the job more thoroughly if you consider me as a candidate. You can see my profile at LinkedIn or Facebook, my name is Kent Munthe Caspersen.

Hi, I have done tweets analysis previously with Python, I think I can help you. As far as I understand the input would be the list of offensive words and a set of tweets, which needs to be processed, or do I need to collect tweets from a particular twitter account?

My name is Mike and I’m from UK. I work with individual clients and also provide outsourcing services for a number of UK and USA based agencies. Your project description sounds interesting to me and I do have skills & experience that are required to complete this project. I can show you some examples of my work. Please contact me to discuss your project.

Hi, I have experience in natural language processing with NLTK and extensive Python experience as well. I would be happy to work with you on this project.

Hello, My name is Irshad, I am expert in python language, I have 5 years+ experience in python language and I am certified python expert. i have completed 100+ projects using python , i have previously worked on programmes that detect OFFENSIVE words from text ,would love to work on this Expertise : Python, Django, Django-Rest-Framework and many python packages. My key Skills are: Python, AngularJS, Scala, JavaScript, Go, PHP, SQL, HTML, Jython, Perl, CSS. Platforms:Linux, Amazon Web Services (AWS), Google App Engine, Windows, Mac OS X I have completed more than 100+ projects in Python programming. You can test the quality of my work if needed. Thanks.

Hi, I have done work similar to this for one of my webiste. I used python to find some keyword from various social website and classified them according to their subjects (heroes from a video game). Finding and tagging offensive words should not be long to do. I could do this in a couple of days, depending on your overall need for the project. You can check my website to have a sample of my work (mainly for image classification).

Dear You, I am really interested to work on this project, I can start the work now , and can provide the best services from my end. Please come on chat to discuss more about the project. Thanks!

Hi, I am a data analyst with more than 2-year experiences. I have expertise in Natural Language Processing. One of my projects involved detecting positive/negative tone in movie reviews and building web application so that when users input their reviews, the system would indicate their tone. I am happy to assist you with this task and am looking forward to hearing back from you. Best regards, Hana

The project listed is quite interesting and also somehow matches with one of my project. I have developed search engine in Python as course work project for subject Information Retrieval. Technically, the application crawls the web pages and processed all the words as token. On the user end, the user simply enter the words she want to search and the pages are retrieved containing the words as the output. To find out similarity between words and pages, I have used cosine similarity formula. I believe this project also can be completed with similar concepts and programming language. As we have already pre-processed data, the overhead of generating the token words is eliminated. So only we have left is to create and use the algorithm for detect and classifying the offensive words from the sentences. For this, the algorithm could be like the sentence will be read and tokenized so that we can use cosine similarity to find best matches and finally we can set a flag. I would like to request you to contact me for further details. Thank you

I'm willing to work for a minimal ammount/free in order to establish a relationship. I'm willing to contact you outside of Freelancer for more flexible payment policies. Cheers

Using Python programming for data scraping, mining and sorting. Additionally can use sqlite queries.

Hi, I'm an Australian based developer, although I only moved back here from the UK in March! I'm experienced in Python, JavaScript and C# and I'm on Freelancer looking to build my client base for my small web development company. I'm reliable and professional. My bid is tentative as there are a number of questions I have. I've assumed that you have a list of offensive words that you want tweets checked against, with the tweet "flagged" if it is found to contain an offensive word. I have completed a similar task on my blog using Python and machine learning for "sentiment analysis", but freelancer rules forbid me from posting a link at this stage. Additionally: 1) Is the offensive "blacklist" always supplied or do you expect the program to find new ones 2) Is the list comprehensive, i.e. does it include common replacements like replacing 'i' with a '1' and so on or do you need these generated and checked as well? 3) how are the tweets (or whatever the inputs are) supplied to the application - should the application pull these directly from Twitter? 4) where is the script run - all the time on a server somewhere or manually on your local computer? 5) what sort of interface do you want for the script? (i.e. a web page or just a command line tool or a desktop application?) If I can have some clarity on these points then I can provide a firmer bid and hopefully we can work together to produce something that meets your requirements! Thanks, Will

I will build the classifier within a week time. I can give a test program for this if you want to test it.