SlideShare uma empresa Scribd logo
1 de 34
DeepCare Chatbot
Generating answers to customers using a hybrid approach
of
Deep Learning and NLP
Pascal van Kooten
2
Introduction
● First conference; first talk
● MSc Methods & Statistics, Utrecht University
● Innovating @ Jibes Data Analytics for ~3 years
● Working for companies to investigate:
– Blockchain
– Computer Vision
– Machine Learning / NLP
● Human Machine Interaction
3
Gartner’s Hype Cycle 2016
4
And this is where you can find me...
5
Fun projects
● whereami
Uses WiFi signals and machine learning to predict where you are
● deep_eye2mouse
Move the mouse by your webcam + gaze
● sky
Using machine intelligence for web crawling
● Neural Complete (presentation in 2 weeks in Moscow)
Neural network to auto complete neural network code
● And…. DeepCare
github.com/kootenpv
6
Chatbots
7
Chatbots
8
Chatbots
● These chatbots take the initiative
● AT&T experimented with chatbots, but cannot
see one online
● Fun experiment to have a reactive chatbot
9
Data
● Find “big QA data” in the English language
● @ATTCares, 1.3M tweets as of May 2017
● 100k Q&A collected through twitter api
● ~15k related to “wanting/needing iphone”
10
Problem
● They receive ~3000 tweets per day
● Corporations want an automated solution
● Very versatile problems:
– Information requests
– Complaints
– Ask for change in system
● Bot needs to detect when it should be quiet
● First task: “social selling”
11
Goal
● Without business constraints…
● Try to see if a bot can reliably reply like a human
● Combine NLP and Deep Learning
– NLP for transparency, answer yes/no
– Deep learning for variation in response
12
http://deepcare.online
(available during PyData)
13
Examples (1/2)
Q: I want an iphone for xmas
A: santa heard you, pvkooten order an iphone at att.com/search?
q=iphone & get #rolloverdata asap. ^fakeAT&Tbot t.co/J4c8G3KR1S
Q: can't decide if I want an iphone 7 or samsung note 4
A: we have great deals on both! get both this your lifestyle! check them
out here att.com/search?q=iphone 7 ^fakeAT&Tbot
Q: I want to see a movie
A: I'm not sure if an iphone is what you really want. I can only help you if
you want an iphone.
14
Examples (2/2)
Q: I want to sell an iphone
A: I'm not sure if an iphone is what you really want. I can only help you if you want an
iphone.
Q: I want to buy an iphone
A: we have got what you need, pvkooten grab an iphone w/ #rolloverdata @ att.com/search?
q=iphone today! t.co/J4c8G3KR1S
Q: I need a charger for my iphone
A: we have one for you, pvkooten get a new charger & many other code tweet-@
att.com/search?q=iphone charger for 20%!^fakeAT&Tbot
Q: i want a case for my iphone 6
A: we have got you covered, pvkooten get a fab case @ att.com/search?q=iphone 6 case
%20cover%20 g & dm me for $$$ off! ^fakeAT&Tbot t.co/J4c8G3KR1S
15
It’s not only deep learning
● Logical mistake:
Q: Stop my samsung contract
A: Ok, I stopped your iphone contract
16
Overview
● Pre-processing
– Generic / Twitter related
– Domain specific
● Phase 1: NLP Model
● Phase 2: Deep Learning answer generation
● Post-processing
Preprocessing
Pascal van Kooten
18
Case study
Q: i want a case for my iphone 6
A: we have got you covered, pvkooten get a fab case @
att.com/search?q=iphone 6 case%20cover%20 g & dm me for $
$$ off! ^fakeAT&Tbot t.co/J4c8G3KR1S
is seen after the preprocessing step as:
Q: i want a case for my PRODUCT
A: we have got you covered, NAME get a fab case @
OFFER_LINK & dm me for MONEY_AMOUNT off! INITIALS
WEBCARE_LINK
19
Generic preprocessing
● Name (API provides actual name) NAME→
● Employee name+time AVAILABILITY→
● Tags (@something) TAG→
● Links LINK (experimented)→
● Emoticons
20
Specific preprocessing
● Brand name
● Products
● Competitors
21
Preprocessing
● Preprocessing to help generalize
● Only preprocess things that can be reused in the
domain
● Plus a crucial component to a company: PRODUCT
● ….
NLP Model
Restrict answering
Pascal van Kooten
23
Traditional NLP
● Depending on language model
– Lemmatization (cats → cat)
– Part of Speech (cat → NOUN)
– Dependency tree (I → subject)
✓Allows generalization!
✗Depends on language model getting it right
✗Feature explosion makes it difficult to generalize
✗Creating rules can be time consuming
✓ It’s very transparant
✓ Rules can give guarantees deep learning can’t
24
Model explanation
● spaCy (https://spacy.io/)
● Rule uses:
– Part of Speech
– Lemma
– Negation
For a sentence,
- need/want... is the lemmatized verb
- which has PRODUCT as a nearby child
- but no negation dependency on the same level or above as PRODUCT
- the subject is “I”
- ???
25
Case study
Q: i want a case for my iphone 6
A: we have got you covered, pvkooten get a fab case @ att.com/search?q=iphone
6 case%20cover%20 g & dm me for $$$ off! ^fakeAT&Tbot t.co/J4c8G3KR1S
want (VERB, ROOT) ✓
---- I (PRON, nsubj) ✓
---- case (NOUN, dobj)
-------- a (DET, det)
-------- for (ADP, prep)
------------ iphone (NOUN, pobj) ✓
---------------- my (ADJ, poss)
---------------- 6 (NUM, nummod)
26
Examples
want galaxy not iphone
not iphone ✓ want iphone!
Generative model
Generating answers
Pascal van Kooten
28
Generative neural network model
● Popularized by Google’s seq2seq:
A general-purpose encoder-decoder framework for Tensorflow that can be
used for Machine Translation, Text Summarization, Conversational Modeling,
Image Captioning, and more.
● Map sequence of words to another sequence of words
● Chinese sentence to English sentence
● Given training examples, minimize loss
● Can also be used for creating a chatbot
● Most likely a variation on this model is being used by Google
Translate
29
Generative neural network model
30
Generative neural network model
https://indico.io/blog/sequence-modeling-neuralnets-part1/
31
Code
● tensorflow
● Started generative model code from
Conchylicultor/DeepQA
32
Neural network parameters
● batch_size: 128
● glob_step: 3186
● max_length: 25
● learning_rate: 0.001
● embedding_size: 128
● hidden_size: 512
● num_layers: 2
---------------------
loss: 0.06
time: 24 hours
33
Conclusion
●
It makes small grammar mistakes
●
It can still make mistakes in logic
– Mostly with unseen data
● Not optimal because 2 models, but fun experiment
● Like OpenAI….
●
Shows signs of sentiment:
Q: I want an iphone but my mom says iphones are the devil
A: do not get upset! order an iphone at att.com/search?q=iphone iphones
with…
34
Questions?

Mais conteúdo relacionado

Semelhante a DeepCare Chatbot - Generating answers to customers using a hybrid approach of Deep Learning and NLP

Everything You Were Taught About Java Is Wrong
Everything You Were Taught About Java Is WrongEverything You Were Taught About Java Is Wrong
Everything You Were Taught About Java Is WrongTim Boudreau
 
Scaffolding a legacy app with BDD scenarios using SpecFlow/Cucumber (HUSTEF 2...
Scaffolding a legacy app with BDD scenarios using SpecFlow/Cucumber (HUSTEF 2...Scaffolding a legacy app with BDD scenarios using SpecFlow/Cucumber (HUSTEF 2...
Scaffolding a legacy app with BDD scenarios using SpecFlow/Cucumber (HUSTEF 2...Gáspár Nagy
 
Limits of Machine Learning
Limits of Machine LearningLimits of Machine Learning
Limits of Machine LearningAlexey Grigorev
 
Mobile App Feature Configuration and A/B Experiments
Mobile App Feature Configuration and A/B ExperimentsMobile App Feature Configuration and A/B Experiments
Mobile App Feature Configuration and A/B Experimentslacyrhoades
 
Pretotyping g motta agile brazil 2013
Pretotyping g motta agile brazil 2013Pretotyping g motta agile brazil 2013
Pretotyping g motta agile brazil 2013Guilherme Motta
 
Cloud Apps - Running Fully Distributed on Mobile Devices - Dominik Rüttimann
Cloud Apps - Running Fully Distributed on Mobile Devices - Dominik RüttimannCloud Apps - Running Fully Distributed on Mobile Devices - Dominik Rüttimann
Cloud Apps - Running Fully Distributed on Mobile Devices - Dominik Rüttimanndistributed matters
 
The case for Web components - Drupal4Gov webinar
The case for Web components - Drupal4Gov webinarThe case for Web components - Drupal4Gov webinar
The case for Web components - Drupal4Gov webinarbtopro
 
Inteligencia artificial para android como empezar
Inteligencia artificial para android como empezarInteligencia artificial para android como empezar
Inteligencia artificial para android como empezarIsabel Palomar
 
Intro To Django
Intro To DjangoIntro To Django
Intro To DjangoUdi Bauman
 
Welcome to a new reality - DeepCrawl Webinar 2018
Welcome to a new reality - DeepCrawl Webinar 2018Welcome to a new reality - DeepCrawl Webinar 2018
Welcome to a new reality - DeepCrawl Webinar 2018Bastian Grimm
 
You shouldneverdo
You shouldneverdoYou shouldneverdo
You shouldneverdodaniil3
 
Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014
Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014
Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014Austin Ogilvie
 
200,000 Lines Later: Our Journey to Manageable Puppet Code
200,000 Lines Later: Our Journey to Manageable Puppet Code200,000 Lines Later: Our Journey to Manageable Puppet Code
200,000 Lines Later: Our Journey to Manageable Puppet CodeDavid Danzilio
 
De-Google Your Life
De-Google Your LifeDe-Google Your Life
De-Google Your LifeLorin Olsen
 
Prepare for the Mobilacalypse
Prepare for the MobilacalypsePrepare for the Mobilacalypse
Prepare for the MobilacalypseJeff Eaton
 
Upwork time log and difficulty 20160523
Upwork time log and difficulty 20160523Upwork time log and difficulty 20160523
Upwork time log and difficulty 20160523Sharon Liu
 

Semelhante a DeepCare Chatbot - Generating answers to customers using a hybrid approach of Deep Learning and NLP (20)

Everything You Were Taught About Java Is Wrong
Everything You Were Taught About Java Is WrongEverything You Were Taught About Java Is Wrong
Everything You Were Taught About Java Is Wrong
 
Scaffolding a legacy app with BDD scenarios using SpecFlow/Cucumber (HUSTEF 2...
Scaffolding a legacy app with BDD scenarios using SpecFlow/Cucumber (HUSTEF 2...Scaffolding a legacy app with BDD scenarios using SpecFlow/Cucumber (HUSTEF 2...
Scaffolding a legacy app with BDD scenarios using SpecFlow/Cucumber (HUSTEF 2...
 
Limits of Machine Learning
Limits of Machine LearningLimits of Machine Learning
Limits of Machine Learning
 
Mobile App Feature Configuration and A/B Experiments
Mobile App Feature Configuration and A/B ExperimentsMobile App Feature Configuration and A/B Experiments
Mobile App Feature Configuration and A/B Experiments
 
Pretotyping g motta agile brazil 2013
Pretotyping g motta agile brazil 2013Pretotyping g motta agile brazil 2013
Pretotyping g motta agile brazil 2013
 
Cloud Apps - Running Fully Distributed on Mobile Devices - Dominik Rüttimann
Cloud Apps - Running Fully Distributed on Mobile Devices - Dominik RüttimannCloud Apps - Running Fully Distributed on Mobile Devices - Dominik Rüttimann
Cloud Apps - Running Fully Distributed on Mobile Devices - Dominik Rüttimann
 
The case for Web components - Drupal4Gov webinar
The case for Web components - Drupal4Gov webinarThe case for Web components - Drupal4Gov webinar
The case for Web components - Drupal4Gov webinar
 
Inteligencia artificial para android como empezar
Inteligencia artificial para android como empezarInteligencia artificial para android como empezar
Inteligencia artificial para android como empezar
 
Introduction to python
Introduction to pythonIntroduction to python
Introduction to python
 
Intro To Django
Intro To DjangoIntro To Django
Intro To Django
 
Welcome to a new reality - DeepCrawl Webinar 2018
Welcome to a new reality - DeepCrawl Webinar 2018Welcome to a new reality - DeepCrawl Webinar 2018
Welcome to a new reality - DeepCrawl Webinar 2018
 
You shouldneverdo
You shouldneverdoYou shouldneverdo
You shouldneverdo
 
Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014
Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014
Applied Data Science: Building a Beer Recommender | Data Science MD - Oct 2014
 
200,000 Lines Later: Our Journey to Manageable Puppet Code
200,000 Lines Later: Our Journey to Manageable Puppet Code200,000 Lines Later: Our Journey to Manageable Puppet Code
200,000 Lines Later: Our Journey to Manageable Puppet Code
 
Keynote 2016
Keynote 2016Keynote 2016
Keynote 2016
 
De-Google Your Life
De-Google Your LifeDe-Google Your Life
De-Google Your Life
 
Prepare for the Mobilacalypse
Prepare for the MobilacalypsePrepare for the Mobilacalypse
Prepare for the Mobilacalypse
 
First android app for workshop using android studio
First android app for workshop using android studio First android app for workshop using android studio
First android app for workshop using android studio
 
Upwork time log and difficulty 20160523
Upwork time log and difficulty 20160523Upwork time log and difficulty 20160523
Upwork time log and difficulty 20160523
 
Big data made easy with a Spark
Big data made easy with a SparkBig data made easy with a Spark
Big data made easy with a Spark
 

Último

Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...ssuserf63bd7
 
GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]📊 Markus Baersch
 
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一F La
 
Semantic Shed - Squashing and Squeezing.pptx
Semantic Shed - Squashing and Squeezing.pptxSemantic Shed - Squashing and Squeezing.pptx
Semantic Shed - Squashing and Squeezing.pptxMike Bennett
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改yuu sss
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDRafezzaman
 
Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)Cathrine Wilhelmsen
 
IMA MSN - Medical Students Network (2).pptx
IMA MSN - Medical Students Network (2).pptxIMA MSN - Medical Students Network (2).pptx
IMA MSN - Medical Students Network (2).pptxdolaknnilon
 
Identifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population MeanIdentifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population MeanMYRABACSAFRA2
 
Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 2Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 217djon017
 
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...Amil Baba Dawood bangali
 
Multiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfMultiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfchwongval
 
MK KOMUNIKASI DATA (TI)komdat komdat.docx
MK KOMUNIKASI DATA (TI)komdat komdat.docxMK KOMUNIKASI DATA (TI)komdat komdat.docx
MK KOMUNIKASI DATA (TI)komdat komdat.docxUnduhUnggah1
 
Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesConf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesTimothy Spann
 
modul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptxmodul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptxaleedritatuxx
 
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档208367051
 
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degreeyuu sss
 
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhh
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhhThiophen Mechanism khhjjjjjjjhhhhhhhhhhh
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhhYasamin16
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfBoston Institute of Analytics
 
Top 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In QueensTop 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In Queensdataanalyticsqueen03
 

Último (20)

Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
 
GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]GA4 Without Cookies [Measure Camp AMS]
GA4 Without Cookies [Measure Camp AMS]
 
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
办理(Vancouver毕业证书)加拿大温哥华岛大学毕业证成绩单原版一比一
 
Semantic Shed - Squashing and Squeezing.pptx
Semantic Shed - Squashing and Squeezing.pptxSemantic Shed - Squashing and Squeezing.pptx
Semantic Shed - Squashing and Squeezing.pptx
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
 
Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)
 
IMA MSN - Medical Students Network (2).pptx
IMA MSN - Medical Students Network (2).pptxIMA MSN - Medical Students Network (2).pptx
IMA MSN - Medical Students Network (2).pptx
 
Identifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population MeanIdentifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population Mean
 
Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 2Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 2
 
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
 
Multiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfMultiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdf
 
MK KOMUNIKASI DATA (TI)komdat komdat.docx
MK KOMUNIKASI DATA (TI)komdat komdat.docxMK KOMUNIKASI DATA (TI)komdat komdat.docx
MK KOMUNIKASI DATA (TI)komdat komdat.docx
 
Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming PipelinesConf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
Conf42-LLM_Adding Generative AI to Real-Time Streaming Pipelines
 
modul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptxmodul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptx
 
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
原版1:1定制南十字星大学毕业证(SCU毕业证)#文凭成绩单#真实留信学历认证永久存档
 
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
 
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhh
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhhThiophen Mechanism khhjjjjjjjhhhhhhhhhhh
Thiophen Mechanism khhjjjjjjjhhhhhhhhhhh
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
 
Top 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In QueensTop 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In Queens
 

DeepCare Chatbot - Generating answers to customers using a hybrid approach of Deep Learning and NLP

  • 1. DeepCare Chatbot Generating answers to customers using a hybrid approach of Deep Learning and NLP Pascal van Kooten
  • 2. 2 Introduction ● First conference; first talk ● MSc Methods & Statistics, Utrecht University ● Innovating @ Jibes Data Analytics for ~3 years ● Working for companies to investigate: – Blockchain – Computer Vision – Machine Learning / NLP ● Human Machine Interaction
  • 4. 4 And this is where you can find me...
  • 5. 5 Fun projects ● whereami Uses WiFi signals and machine learning to predict where you are ● deep_eye2mouse Move the mouse by your webcam + gaze ● sky Using machine intelligence for web crawling ● Neural Complete (presentation in 2 weeks in Moscow) Neural network to auto complete neural network code ● And…. DeepCare github.com/kootenpv
  • 8. 8 Chatbots ● These chatbots take the initiative ● AT&T experimented with chatbots, but cannot see one online ● Fun experiment to have a reactive chatbot
  • 9. 9 Data ● Find “big QA data” in the English language ● @ATTCares, 1.3M tweets as of May 2017 ● 100k Q&A collected through twitter api ● ~15k related to “wanting/needing iphone”
  • 10. 10 Problem ● They receive ~3000 tweets per day ● Corporations want an automated solution ● Very versatile problems: – Information requests – Complaints – Ask for change in system ● Bot needs to detect when it should be quiet ● First task: “social selling”
  • 11. 11 Goal ● Without business constraints… ● Try to see if a bot can reliably reply like a human ● Combine NLP and Deep Learning – NLP for transparency, answer yes/no – Deep learning for variation in response
  • 13. 13 Examples (1/2) Q: I want an iphone for xmas A: santa heard you, pvkooten order an iphone at att.com/search? q=iphone & get #rolloverdata asap. ^fakeAT&Tbot t.co/J4c8G3KR1S Q: can't decide if I want an iphone 7 or samsung note 4 A: we have great deals on both! get both this your lifestyle! check them out here att.com/search?q=iphone 7 ^fakeAT&Tbot Q: I want to see a movie A: I'm not sure if an iphone is what you really want. I can only help you if you want an iphone.
  • 14. 14 Examples (2/2) Q: I want to sell an iphone A: I'm not sure if an iphone is what you really want. I can only help you if you want an iphone. Q: I want to buy an iphone A: we have got what you need, pvkooten grab an iphone w/ #rolloverdata @ att.com/search? q=iphone today! t.co/J4c8G3KR1S Q: I need a charger for my iphone A: we have one for you, pvkooten get a new charger & many other code tweet-@ att.com/search?q=iphone charger for 20%!^fakeAT&Tbot Q: i want a case for my iphone 6 A: we have got you covered, pvkooten get a fab case @ att.com/search?q=iphone 6 case %20cover%20 g & dm me for $$$ off! ^fakeAT&Tbot t.co/J4c8G3KR1S
  • 15. 15 It’s not only deep learning ● Logical mistake: Q: Stop my samsung contract A: Ok, I stopped your iphone contract
  • 16. 16 Overview ● Pre-processing – Generic / Twitter related – Domain specific ● Phase 1: NLP Model ● Phase 2: Deep Learning answer generation ● Post-processing
  • 18. 18 Case study Q: i want a case for my iphone 6 A: we have got you covered, pvkooten get a fab case @ att.com/search?q=iphone 6 case%20cover%20 g & dm me for $ $$ off! ^fakeAT&Tbot t.co/J4c8G3KR1S is seen after the preprocessing step as: Q: i want a case for my PRODUCT A: we have got you covered, NAME get a fab case @ OFFER_LINK & dm me for MONEY_AMOUNT off! INITIALS WEBCARE_LINK
  • 19. 19 Generic preprocessing ● Name (API provides actual name) NAME→ ● Employee name+time AVAILABILITY→ ● Tags (@something) TAG→ ● Links LINK (experimented)→ ● Emoticons
  • 20. 20 Specific preprocessing ● Brand name ● Products ● Competitors
  • 21. 21 Preprocessing ● Preprocessing to help generalize ● Only preprocess things that can be reused in the domain ● Plus a crucial component to a company: PRODUCT ● ….
  • 23. 23 Traditional NLP ● Depending on language model – Lemmatization (cats → cat) – Part of Speech (cat → NOUN) – Dependency tree (I → subject) ✓Allows generalization! ✗Depends on language model getting it right ✗Feature explosion makes it difficult to generalize ✗Creating rules can be time consuming ✓ It’s very transparant ✓ Rules can give guarantees deep learning can’t
  • 24. 24 Model explanation ● spaCy (https://spacy.io/) ● Rule uses: – Part of Speech – Lemma – Negation For a sentence, - need/want... is the lemmatized verb - which has PRODUCT as a nearby child - but no negation dependency on the same level or above as PRODUCT - the subject is “I” - ???
  • 25. 25 Case study Q: i want a case for my iphone 6 A: we have got you covered, pvkooten get a fab case @ att.com/search?q=iphone 6 case%20cover%20 g & dm me for $$$ off! ^fakeAT&Tbot t.co/J4c8G3KR1S want (VERB, ROOT) ✓ ---- I (PRON, nsubj) ✓ ---- case (NOUN, dobj) -------- a (DET, det) -------- for (ADP, prep) ------------ iphone (NOUN, pobj) ✓ ---------------- my (ADJ, poss) ---------------- 6 (NUM, nummod)
  • 26. 26 Examples want galaxy not iphone not iphone ✓ want iphone!
  • 28. 28 Generative neural network model ● Popularized by Google’s seq2seq: A general-purpose encoder-decoder framework for Tensorflow that can be used for Machine Translation, Text Summarization, Conversational Modeling, Image Captioning, and more. ● Map sequence of words to another sequence of words ● Chinese sentence to English sentence ● Given training examples, minimize loss ● Can also be used for creating a chatbot ● Most likely a variation on this model is being used by Google Translate
  • 30. 30 Generative neural network model https://indico.io/blog/sequence-modeling-neuralnets-part1/
  • 31. 31 Code ● tensorflow ● Started generative model code from Conchylicultor/DeepQA
  • 32. 32 Neural network parameters ● batch_size: 128 ● glob_step: 3186 ● max_length: 25 ● learning_rate: 0.001 ● embedding_size: 128 ● hidden_size: 512 ● num_layers: 2 --------------------- loss: 0.06 time: 24 hours
  • 33. 33 Conclusion ● It makes small grammar mistakes ● It can still make mistakes in logic – Mostly with unseen data ● Not optimal because 2 models, but fun experiment ● Like OpenAI…. ● Shows signs of sentiment: Q: I want an iphone but my mom says iphones are the devil A: do not get upset! order an iphone at att.com/search?q=iphone iphones with…

Notas do Editor

  1. - Share a project of mine, I’m enthousiastic about - Thank you for coming - Did you come for deep learning? - NLP? - Chatbot?
  2. I really like it at Jibes, if you’re living in the Netherlands and you’re passionate about python and data…. Contact us
  3. First web Then phone apps Now chatbots, in e.g. facebook messenger Or on website Powerful:
  4. First web Then phone apps Now chatbots, in e.g. facebook messenger Or on website
  5. Tweets through API API
  6. Tweets through API API
  7. “Full stack” chatbot Capable of replying to tweets…. → talking you into buying an iphone I want an iphone for xmas Backend in Python Frontend in Angular 2
  8. - Markov chain - I would like to eat - I would like to drink
  9. - Find out what globstep is - twitter limited to 140 characters - this data set ~25 tokens -