Machine Learning: Document Classificiation

Hello! I'm a computer engineer and a Phd student. My research focus is on data science and machine learning. I have previously implemented prediction and recommendation systems in different domains such as bioinformatics and social networks. In another project, I've worked on classifiying academic papers to topics based on the content of the paper. In that project, I designed a system which learned the words that are strong indicator of a specific topic and classified the papers according to relevance of these words. I think we can use similar strategy to solve your problem. Looking forward to talking to you! - ykocak

A proposal has not yet been provided

I am a Mathfin graduate and have done such projects in academic and professional setting. I have fair amount of experience in machine learning especially MLP and have experience in using nlp packages in Python as NLTK, Gensim. I am really proficient in Python and R and can do these implementations with rigorous detail. Let me know more about the task, so that I can take a look

I'm an active competitive programmer. I have a lot of experience with algorithmic and machine learning tasks.

My main field of interest at the moment is machine learning and not so long ago I have worked intesively with extracting information from PDFs as well as editing them. Depending on the data you have, I am thinking of one of two strategies: If you have labeled PDFs, already classified, we can use them almost immediately with some preprocessing to train a classifier. If not, we have to use some natural language processing techniques to extract specific features that will help in the classification. I am supposing, of course, that the classification will be done on text data from the PDFs

I am a programmer in a range of technical areas involving Web, IoT, BI and BD Analytics. I primarily work in Java and Java related technologies, however I have been working in a variety of programming languages such as Python, JavaScript, Scala. My area of expertise is in Analytics for Web, IoT and BI, involving the new age technologies for development of real time analytics and complex algorithm on a wide variety of dimensions of data. I am proficient in ML, NLP frameworks that are available in the open source world, and has around 9.5 years of experience in the industry. I see that you have challenging requirement and an opportunity to work together would be ideal. I wish to talk to you more about this and would appreciate the opportunity to discuss with you how my assistance would drive your goals to success.

I'm a very tallented programmer with great algorythmic skills which is proved by many programming contest I have won (they are listed in my profile description). I know neural networks techniques and used them successfully to do image recognition in the past. I guess this time PDFs are to be recognised by the text they they contain so I would use word2vec first and then apply either RNN or CNN depending on which one gives better results. I could say more if I saw a few examples of PDFs.

Approach to organize (Document clustering)and name PDF files Reading PDF File contents. Using JAVA language, APACHE PDF Box Open Source tool Indexing Each file. Using Java to index files for classification Tagging document type based on contents. Using best suited methods for data (Naive Bayes Classification, aximum Entropy Classification, or probabilistic Grammar Classification) Naming document as per relevant content inside. Java using classified data For more details Please feel free to contact. You will like my work and dedication on this

statistics - data mining / machine learning, pattern recognition, neural network, random forest, ridge regression, lasso, nearest neighbor, cluster analysis, multiple linear regression, logistic regression, logit / probit transformation, spline, kernel smoother, nonparametric statistics, support vector machines, cross-validation, model selection, principal component analysis ( PCA ), canonical correlation analysis, Monte Carlo, variance reduction, antithetic / importance sampling, covariate, Markov Chain Monte Carlo, EM algorithm, Gibbs sampler, Metropolis - Hastings, Bayes rule, Bayesian Statistics, conjugate prior, posterior distribution, conditional expectation, multivariate distribution,