Categorize text documents

[muejan]

Hi everybody,

so far I have been a quite basic user of KNIME, done some easy stuff and I know well about all the general usage of KNIME like building workflows etc. Now I wanted to dig a bit deeper and try out more of the incredible capabilities.

In fact I have a bunch of text files, that are all “categorized”, means they have different topics assigned, one topic for each document.

I now want to train a model with the existing files, and let KNIME then use the model to classify a bunch of untagged files.

I tried to find a good starting point to learn how to do it, but to be honest my research gave me so many results, different plugins and situations I was kinda lost what fits my problem.

Would you experts have any recommendation where to start from? This would be really highly appreciated!
(I don’t expect any solution, a link or tut would be enough for me to learn. Yet also if theres kinda blueprint or soemthing this still would be nice…)

Thanks a lot in advance!

Best wishes,
Jo

[qqilihq]

Hi Jo,

here’s two pointers to existing threads which should help you to get started:

Text classification starting from excel

I have an excel file consisting of 700 rows and 2 columns. The first column contains the category, the second the contents of an email message. I want to use this file to train a naive Baysian learner to classify another file with new input (unclassified). What are the minimum components I need (apart from the file reader and the Naive Bayes nodes)? How to connect and set them up? Any help is greatly appreciated! Peter.

Classifying Documents doesn't work using Naive Bayes

@josifrs sorry that it took me so long to get back to you. While investigating your problem we came across a corner case that KNIME’s Naive Bayes Predictor handles slightly differently to other implementations and this took some time to adapt (please download 3.7.1 as once it is released - should be very soon). Anyway, any other Naive Bayes implementation I tested always predicts the same class, or nothing at all . I’ll spare you the details. You could try another classifier, a…

Let us know how it goes!

Philipp

[ScottF]

First of all, you will want to be familiar with the KNIME Textprocessing extension. We have recently released a series of short videos on our KNIMETV Youtube Channel to help familiarize folks with the extension:

You can also find several examples of workflows for document classification on the Hub. Here’s one such example:

Sentiment Analysis (Classification) of Documents – kilian.thiel

[Daniel_Weikert]

Was a great series. Would love to see more on the KNIME channel

[muejan]

Dear Philip, dear Scott,

this seems very helpful and helps me a lot to narrow down the points where to start from!
Thanks a lot, highly appreciate it!