URL Path Analysis

I am a machine learning guy, who has Ph.D. in machine learning and natural language processing. I am also a kaggler, once listed in topo 500 among the world wide kagglers. Have a look at my profile: [login to view URL] There are two approaches I can come up with. First, I can try only using information provided within the URL and build a classification model, if the model is accurate enough this would be the best cause it runs much faster. Second, I can crawl pages pointed by the URL and do webpage classification. This model will be accurate but might crawling time dominate. Lets discuss further! Best Tsendee

Hi, Information Security expert is here. So, you need a machine learning algorithm to perform the analysis. I can do this job. Please come in chat for more info. Thank You.

Hello. I take an interest in your job. I have full skills in Web scrapping, Android, Iphone, Linux, Reverse engineering, Java, AI, Mathematics, Algorithms, Qt, c++ & c# programming. Please check my profile and portfolio. I'd be happy to keep long partnership,if you'd like. Looking forward for your kind answer. Best Regards. Yknox.

Hi I am a very experienced statistician and have worked with several companies as a analyst. I will apply classification algorithms like PCA , Randomforest, KNN or LDA to choose the best classifier algorithm for the data and will provide high efficiency in classification of risky Urls from safe. Looking forward to discussion, Best Regards, Suyash

Hi I have done several projects on machine learning. Your task is essentially one of the text classification. We can use SVM for classification and use Bag of Words approach for extracting features, and then build on that. Kindly share more details. Regards

I am adept at using machine learning algorithms like random forest, xgboost, ada boost and neural networks in R. If your task is to classify urls as good or bad /risky, I can help you build a model which will automatically do this.

Respected Sir/Madam, Thanks for taking interest in our bid. ______________________________________________________________________________________________________________ Short description: OS : Macintosh, Windows, Android and Linux (32/64 bit) Editor : Net-beans & Eclipse Technology : JAVA, Mysql, SQL, HTML, CSS, Javascript. ______________________________________________________________________________________________________________ About Us: We have 8+ years experience in Java software development. We have successfully developed 260+ projects and Games, Image processing, Network programming, Web development, mobile apps, etc. ______________________________________________________________________________________________________________ Our software solutions: We are working in NetBeans IDE 8.0 and Eclipse. We will develop your project with QA and testing, also write technical comments in source code, So you can easily understand the flow of project. _____________________________________________________________________________________________________________ Before awarding the project please clarify, 1) Deadline. 2) All requirements with necessary documents. 3) And are you an owner or a third party of this project? _____________________________________________________________________________________________________________ Contact Us: Drop the massage over here. We are mostly on line.

Bir öneri henüz sağlanmadı

This sounds like a machine learning problem that needs a classifier implemented and trained. I will need more info for a clear judgement. I have a bachelor in CS from the American University in Cairo and a minor in Mathematics. I have worked for the past year in Microsoft's Advanced Technology Lab in Cairo (ATLC) as a research assistant in the Machine Learning team. I have worked on various machine learning problems. Additionally, I am an IEEE published author.

★★ Machine learning expert and software architect based in the United States - feel free to message me to discuss requirements prior to the project being awarded. ☞ How I Can Help You: I'm guessing you want to use these URLs to train a classifier in order to be able to evaluate new URLs in terms of their likelihood of being dangerous. It sounds easier than it is. We can definitely crawl the websites, capture whois information, dns server info, location of the physical box in the world, etc...The hard part becomes figuring out which of these features help in determining the trustworthiness of a URL. For example, I'd assume location to be a big one. Shoot me a message so we can discuss your specific requirements and plan. Afterwards I will propose a set of milestones. ☞ About Me: My name is Tom Griffin, I'm based in the United States, and I have fifteen years of experience in a variety of technology and business roles. You may read more about me on LinkedIn if you search for "Tom Griffin IEEE." ☞ Warranty: All of my work comes with a 30-day warranty in the event something fails to work or experiences issues. ║✭✭✭✭✭✭✭▅▅▅▅▅▅▅▅▅ Tom is based in the United States ║✭✭✭✭✭✭✭▅▅▅▅▅▅▅▅▅ and has worked with major organizations. ║✭✭✭✭✭✭✭▅▅▅▅▅▅▅▅▅ You can read more about him in his profile ║▅▅▅▅▅▅▅▅▅▅▅▅▅▅▅ or by finding him on LinkedIn by searching for ║▅▅▅▅▅▅▅▅▅▅▅▅▅▅▅ "Tom Griffin IEEE." ║▅▅▅▅▅▅▅▅▅▅▅▅▅▅▅

What is your Goal after this analysis? Just try to apply the ML to ur Data or is there any specific Goal after ML output Any planned ML algorithms in ur mind ? How u r going to create engine ( Supervised / Un supervised) ? Do you have any idea to merge diff ML algorithms?

Currently a student of NYU pursuing MS in Comp Sci with Machine Learning as my field of concentration. Have worked with a lot of Machine Learning algorithms for Natural Language Processing. Used PredictionIO for automatic product categorization of fashion products for a e-commerce startup. Also worked with BioNLP data and SVMs for relation extraction and event extraction. With my experience in Machine Learning I think I am the best suited for this project. I have worked with various kinds of data and I am sure I can show you results.

Предложение еще не подано