Text Mining Use Cases and Capabilities with RapidMiner

yyhuang

Attached is the slide deck that summarizes the major functions and techniques offered by the Text Processing, Web Mining, and Operator Toolbox extensions.

Telcontar120

@yyhuanglooks like some great new applications here. The presentation contains screenshots that appear to be of annotated processes---are there any new templates or examples that correspond to some of the newer capabilities described in the deck available for download?

Thomas_Ott

Thanks@yyhuang, this has inspired me to upgrade my Twitter Content models.

yyhuang

Thanks for your interests! Of course, I will share some template process as supplemental files to the slides.

Telcontar120

@yyhuangany ETA on those template processes? :-)

Thanks!

yyhuang

<连接from_op = "过滤器Stopwords(英语)”来m_port="document" to_op="Stem (WordNet)" to_port="document"/>























LSA is quite simple, you just use SVD to perform dimensionality reduction on the tf-idf vectors&#8211;that&#8217;s really all there is to it!

LSA or LSI

yyhuang

generate n-grams with max length of 5

LDA

The used filter on token is a data like

and
the
you
for
in
on
from
of
am
is
was
are
be
i
that
with
very
really
can
has
will
this
they

yyhuang

<运营商激活= " true "类=“子流程”薪酬atibility="8.1.003" expanded="true" height="82" name="Get News Feeds" width="90" x="45" y="187">


http://feeds.bbci.co.uk/news/rss.xml"/>


http://feeds.bbci.co.uk/news/world/asia/rss.xml"/>


http://feeds.bbci.co.uk/news/business/rss.xml"/>


http://feeds.bbci.co.uk/news/entertainment_and_arts/rss.xml"/>









Don't convert article link to document text.




















































We use a common process for both the training and testing set



<运营商激活= " true "类=“子流程”薪酬atibility="8.1.003" expanded="true" height="103" name="post-processing" width="90" x="1184" y="85">

<操作符= " true " class = " generate_attribu激活tes" compatibility="8.1.003" expanded="true" height="82" name="Generate Attributes (4)" width="90" x="45" y="34">










<操作符= " true " class = " generate_attribu激活tes" compatibility="8.1.003" expanded="true" height="82" name="Generate Attributes (2)" width="90" x="246" y="34">














<操作符= " true " class = " generate_attribu激活tes" compatibility="8.1.003" expanded="true" height="82" name="Generate Attributes (3)" width="90" x="514" y="34">










<连接from_op = from_po“生成属性(2)”rt="example set output" to_op="Select Attributes (2)" to_port="example set input"/>

<连接from_op = from_po“生成属性(3)”rt="example set output" to_op="Rename" to_port="example set input"/>







































































Extracted compnay name and find co-existence<br>find similar documents with the metioned entity name

Entity recoginition for the company names mentioned in news tittle, with a lisf of target like

19_entertainment
20th_century_fox
23andme
27b/6
37signals
3com
3m
7-eleven
a&m_records
a&w_root_beer

Telcontar120

Wonderful, thanks so much for those!!