Text Mining by Example in SAS® Enterprise Miner™

[#1]

Download the Files (GitHub)

This tip is part of Learn by Example using SAS® Enterprise Miner^TM series where a new data mining topic is introduced and explained with one or more example SAS Enterprise Miner process flow diagrams.

Text Mining is about extracting relevant information from a collection of text documents to uncover the underlying themes and concepts. The integration of SAS Text Miner nodes in a SAS Enterprise Miner process flow diagram enables you to combine quantitative variables with unstructured text thus incorporating text mining with other data mining techniques.

SAS Text Miner supports an extensive list of languages, refer to the product page for additional details. Note that to run these examples, you will require SAS Text Miner add-on to SAS Enterprise Miner installation.

To get started, download the process flow diagrams (XML files) and the accompanying PDF documentation for the following two examples from the GitHub repository at https://github.com/sassoftware/dm-flow/tree/master/TextMining

1. Text Mining Explore: This example uses different techniques to explore textual data or articles. The Text Parsing node takes the raw text and quantifies the terms therein; the Text Filter node filters out any extraneous information; the Text Rule Builder node generates an ordered set of rules that are useful in describing/predicting the target; the Text Cluster node uses SVD or Singular Value Decomposition to cluster articles into multiple groups and the Text Topic node extracts the topics from the article collection. The Control Point node at the end enables the execution of the three flows in this example with a single click.

2. Text Mining Classification: This example classifies textual articles into a news group (graphics, hockey or medical) based on their content. Using a similar flow as in above example, the topics are first extracted and that information is subsequently used in the classification model (Regression, Neural Network, Decision Tree and Memory Based Reasoning) to pick a champion. Finally, the SAS Code node is used to print the list of articles and their predicted news groups in to which they are classified.

To run these examples, refer to the README file that is part of the GitHub repository at https://github.com/sassoftware/dm-flow. Please note that these examples were tested with SAS Enterprise Miner 13.2 and SAS Text Miner 13.2.

[Prajna_450]

hi guys,

I am trying to learn sas text mining. I need help to understand the tool with some vedios probably.

Kindly guide if there are any webinars recorded or suggest a book if available

Thank you,

Prajna

[RadhikhaMyneni]

Hi Prajna,

For videos, you can go onto youtube and search for "sas text mining" which you can use to learn visually. There is also a book on applications and use cases on this topic here Text Mining and Analysis

There is a new learning center that came live recently on Enterprise Miner where you will find lot of free learning resources on the product. You can get to it at https://www.sas.com/en_us/learn/software/enterprise-miner.html

Hope this helps,

Radhikha

[Prajna_450]

Hi Radika,

Thank you for yor response, are there any material for sentiment anlysis in
sas eminer?

Please helpout!

Thank you

##- Please type your reply above this line. Simple formatting, no
attachments. -##

[Prajna_450]

Hi,

To obtain text sentiment scores will I need to download sas sentiment Analysis studio ? if yes, is there a trail version too for the same ?

thank you,

Prajna