Discover the new MeaningCloud Extension for RapidMiner.
MeaningCloud webinar, April 27th, 2017.
More information and contents of the webinar https://www.meaningcloud.com/blog/recorded-webinar-integrate-the-most-advanced-text-analytics-into-your-predictive-models
www.meaningcloud.com
2. MeaningCloud Extension for RapidMiner
Before we get started…
Presenter
How to participate
• Send questions with the chat feature, or
• Click the “Raise your hand” button to speak
and we’ll enable your mic
• Afterwards, you’ll be able to access a recording of the
webinar and its contents as tutorials on our blog
Antonio Matarranz
CMO
3. MeaningCloud Extension for RapidMiner
The purpose of this webinar…
To learn how to combine
text and data in advanced
analytical models
4. MeaningCloud Extension for RapidMiner
Agenda
Analytics platforms. Introduction to RapidMiner
Text analytics. Introduction to MeaningCloud
Combining text and data analytics. MeaningCloud
Extension for RapidMiner
Practical case demo
Application scenarios
How this Extension is different
Product roadmap
Conclusions and Q&A
5. MeaningCloud Extension for RapidMiner
Data Prep
Speed and optimize all data
exploration, blending and
cleansing tasks
Operationalize
Easily deploy and
maintain models and
embed analytic results
Model & Validate
Apply machine learning to rapidly
prototype and confidently
validate predictive models
Embed results in
all types of
business apps and
data visualization
tools
Incorporate all
types of data
ACCELERATES TIME TO VALUE
Integrated analytics platforms
7. MeaningCloud Extension for RapidMiner
RapidMiner Studio
Access to Data and
Processes
All kind of Operators,
including 129 for
Modelling
8. MeaningCloud Extension for RapidMiner
Deploy
• High velocity scoring
• APIs to deploy your predictive
results into the broadest array of
business applications, e.g.,
Salesforce, Marketo, Tableau
• Lightning-fast creation of web-
based reports
• Model monitoring tools to
ensure continued performance
4
Prep Data
• Access all types of data:
structured, unstructured and
Big Data sources
• Interactive data visualizations.
Anomaly & outlier detection
• Normalization & standardization
• Dimension reduction / feature
selection
1
Validate
• Breadth of validation schemes
• Cross-validation
• Visual evaluation
• Honest validation
• Encapsulate data prep,
modeling into validation
• Model performance metrics
• Cluster performance measures
3
Model
• Statistical & machine learning
• Predictive modeling
• Segmentation & clustering
• Association mining
• Similarity computation
• Feature weighting
• Model, parameter & features
optimization
2
Ability to visually execute
your entire Predictive
Analytics workflow in a
single place
Mainly focused on
structured data
RapidMiner Studio
9. MeaningCloud Extension for RapidMiner
Why should we be using text analytics?
Structured data
Unstructured
content
10. MeaningCloud Extension for RapidMiner
Opinions
Facts
Concepts
Organizations
People
Semantic
Analysis
Relationships
Themes
Text analytics
Extract meaning and actionable insights from unstructured content
Automation of costly manual activities
11. MeaningCloud Extension for RapidMiner
MeaningCloud: “Meaning as a Service”
(SaaS and on-premises)
Sign up and use it for FREE at
http://www.meaningcloud.com
12. MeaningCloud Extension for RapidMiner
MeaningCloud’s APIs
Identifies occurrences of
names of people,
organizations, abstract
concepts, quantities, etc.
Theme classification
according to
predefined taxonomies
Identifies general and
attribute-level polarity
Distinguishes among 60
languages
Performs detailed morphosyntactic
analysis
Evaluates the impact of
opinions on several
reputational axes
Discovers meaningful topics
and similarities among texts
without relying on predefined
taxonomies
13. MeaningCloud Extension for RapidMiner
Add-in for Excel
An experience fully integrated into Excel
Easy to use - No programming!
The most convenient way to evaluate, prototype, and use MeaningCloud
13
15. MeaningCloud Extension for RapidMiner
MeaningCloud Extension for RapidMiner
Integrating the most advanced text analytics into RapidMiner
Download it here
16. MeaningCloud Extension for RapidMiner
MeaningCloud Extension for RapidMiner
Combine text analytics and
structured data in powerful
predictive models
Operators for
• Topics Extraction
• Text Classification
• Sentiment Analysis
• Lemmatization
Access to personal
resources created with
Customization Tools
18. MeaningCloud Extension for RapidMiner
Analysis of comments from Amazon
Data: 1,500 food reviews from Amazon (vía Kaggle)
Structured data Unstructured text
19. MeaningCloud Extension for RapidMiner
Questions we would like to ask
Is there any (co)relation between Score and Sentiment?
1 2 3 4 5
Score
P+ → 5
P → 4
Sentiment NEU → 3
N → 2
N+ → 1
20. MeaningCloud Extension for RapidMiner
What (co)relation exists between Score and Sentiment?
Correlation Score – Polarity
21. MeaningCloud Extension for RapidMiner
Questions we would like to ask
Which attributes have the biggest impact on sentiment?
Predictive analytics
Model: factors that
predict sentiment
f(Atr1, Atr2, Atr3,…)
Texto Atr1 Atr2 Atr3 … Sentim
This … 1 0 1 … P
I am … 0 1 1 … N
… … … … … …
Texto Atr1 Atr2 Atr3 … Sentim
Your … 0 1 0 … N
Today… 1 0 0 … P
… … … … … …
Texto Atr1 Atr2 Atr3 … Sentim Pred
Your … 0 1 0 … N N
Today… 1 0 0 … P NEU
… … … … … …
Training set
Test set
22. MeaningCloud Extension for RapidMiner
Which attributes have the biggest impact on sentiment?
Rule model
if HelpfulnessDenominator ≤ 0.500 and con_food ≤ 0.500 then positive (39 / 339 / 39)
if con_product ≤ 0.500 and HelpfulnessNumerator > 1.500 and ent_tea > 0.500 then positive (1 / 25 / 2)
if con_$ ≤ 0.500 and HelpfulnessNumerator > 0.500 and con_mistake ≤ 0.500 then positive (49 / 340 / 52)
if HelpfulnessNumerator ≤ 0.500 and HelpfulnessDenominator ≤ 5 and con_chip ≤ 0.500 and con_restaurant ≤ 0.500 and
con_beef ≤ 0.500 and ent_Science ≤ 0.500 and con_baby ≤ 0.500 and con_world ≤ 0.500 and con_pill ≤ 0.500 and
ent_Food_and_Drug_Administration ≤ 0.500 and ent_HAM_Base ≤ 0.500 and con_consumer ≤ 0.500 and con_book
≤ 0.500 then positive (8 / 74 / 3)
if con_snack > 0.500 then positive (0 / 3 / 0)
if HelpfulnessDenominator > 11.500 and HelpfulnessNumerator > 13 then neutral (3 / 1 / 0)
if HelpfulnessDenominator > 4.500 and HelpfulnessDenominator ≤ 7.500 then negative (0 / 0 / 3)
if con_can > 0.500 and con_scratch ≤ 0.500 then positive (0 / 3 / 0)
if HelpfulnessNumerator ≤ 2.500 and con_baby ≤ 0.500 then neutral (13 / 6 / 10)
if HelpfulnessNumerator ≤ 9 then positive (2 / 7 / 2)
else negative (0 / 0 / 1)
correct: 811 out of 1025 training examples.
23. MeaningCloud Extension for RapidMiner
Which attributes have the biggest impact on sentiment?
Performance Vector
25. MeaningCloud Extension for RapidMiner
Combining data and unstructured information
Customer data
Consumption /
use activity
Interactions /
incidents
Social
comments
Predictive analytics
More actionable
insights
Increased
predictive capacity
Model: factors
that predict
variables
Enriching models purely based on structured data
26. MeaningCloud Extension for RapidMiner
Application scenarios
Root cause analysis Fraud & churn prevention
Segmentation, targeting &
scoring People analytics
27. MeaningCloud Extension for RapidMiner
Opinions
The sentence “The
highest interest rate in
industry!” is…
Positive, if talking
about savings
Negative, if talking
about mortgages
Customized linguistic resources improve accuracy
Mentions
Names of banks and
financial companies,
e.g., JPMorgan, BNP
Paribas, Citibank
Product names, e.g.,
Your Way Account.
Compass Account…
Themes
Example: analysis of bank’s customer opinions
Products
Accounts
Checking
Savings
Borrowing
Credit
Mortgage
Channel
Office
Phone
Internet
28. MeaningCloud Extension for RapidMiner
Customization tools
Create your own dictionaries, classification
models, and sentiment analysis
Graphical user interface - no programming!
Improve precision & recall
Learn more about customization in this webinar
29. MeaningCloud Extension for RapidMiner
A view into the future
Usability
URL parameter
Documents: creation and file
parameter
Language identification
Aspect-based sentiment analysis
PoS (Part of Speech) tagging
Text clustering
User profiling
Vertical packs, e.g., banking, health
Emotion detection
Intent detection
Q2 2017 Q3 2017 Q4 2017 Q1 2018 Q2 2018
Roadmap MeaningCloud Extension for RapidMiner
GA
30. MeaningCloud Extension for RapidMiner
In conclusion
Close integration between
RapidMiner and MeaningCloud
For RapidMiner users
For MeaningCloud users
Data + text combination
boost model value
32. MeaningCloud Extension for RapidMiner
Stay tuned to our emails and blog
We’ll be posting a recording of the webinar and
its contents as tutorials soon
33. MeaningCloud Extension for RapidMiner
Thank you for your attention!
MeaningCloud LLC
54 W. 40th St.
New York, NY 10018
+1 (646) 403-3104
MeaningCloud Europe SL
Llano Castellano 13
28034 Madrid (Spain)
+34 91 3324301
sales@meaningcloud.com
support@meaningcloud.com
http://www.meaningcloud.com
@MeaningCloud
https://www.linkedin.com/company/meaningcloud