SlideShare a Scribd company logo
1 of 39
ABOUT CAMPFIRE
Free PowerPoint
Templates
HotDesk
Fixed DeskPrivate Offices
Event Space rental
Kennedy Town
Free PowerPoint Templates
Wong Chuk Hang
Photo studio
Hot desk
Runway
Fixed Desk Private Offices Event Space rental
Free PowerPoint
Templates
Campfire community events
Contact us
Kennedy Town
k
+852 9228 7163
Machine Learning in
eCommerce
USING MACHINE LEARNING TO INCREASE CONVERSIONS AND SALES
REBOOT.AI
Agenda
 Who: Who are we?
 Why: Why value does machine learning add in eCommerce?
 What: What algorithms are used in eCommerce?
 Introduction to algorithms
 Business Use Cases : User Personas, Product Recommendations
 Which algorithm is used for this specific case?
 How: How does this algorithm actually work?
 High Level Description
 Example Implementation
 Q&A
About Reboot.ai
 Matt O’Connor
 BBA Finance
 Previous: Lead Trader Algorithmic Desk - Macro Hedge Fund
 Current: Full stack developer and professional Scrum Master (PSM I)
 Avid futurist –social ramifications of AI & blockchain
 Dhruv Sahi
 BA Mathematics and Economics
 Previous: Data Science Chief – Smart Cities Startup
 Current: Business Intelligence Analyst – eCommerce - Grana
 AI, IoT, and smart cities enthusiast
 Reboot.ai
 Hong Kong’s only dedicated machine learning and AI training provider
 Part time evening courses for beginners and advanced
 Curriculums developed in partnership with local data companies
 Use ML & AI in our classrooms to improve teaching and personalize learning
Who?
Why Machine Learning?
 1) Computers much faster than humans
 Even complex or infinite solution problems have practical ‘solutions’ and optimizations
 Ex. Google maps vs human intuition
 2) Logic is replicable and scalable
 Consistency of results not humanly possible
 Conducive to experimentation and A/B testing can limit variables at play
 3) Can incorporate elements of ‘learning’ from results
 Can ‘teach itself’ and improve
 Can identify insights that are not intuitive or sometimes invisible to humans
Why?
Headline Use Cases
 Recommendation Engines: How Amazon and Netflix Are Winning the Personalization
Battle and optimizing revenues
 75% of all content on Netflix is viewed through their recommendation engine
 35% of Amazon’s revenues are the product of their recommendation engine
 Machine Learning Generates Clickbait Headlines That Will SHOCK You
 Predict Sentiment From Movie Reviews Using Deep Learning
 Can Chatbots Help Reduce Customer Service Costs by 30%?
Why?
Separating Two Customer Groups…
Why?
…Is Easy for a machine
Why?
Multi Dimensional Reasoning
Why?
Summary: Why?
 Benefits
 Can be faster and cheaper than human alternative
 Can be employed in a wide variety of real world conditions even with limited/flawed data
 Can improve, learn, and identify trends humans would have trouble identifying
 Weaknesses
 Very difficult to create intelligence good in multiple unrelated contexts
 No instincts, ‘genetic knowledge’ or ‘intuition’
 Mistrusted and misunderstood
 Questions?
Why?
What is an Algorithm
 An algorithm is a step by step process for completing a task.
 Everyday examples: recipes, ‘habits’, traditions, traffic laws
 Example in code
emailCustomer(gender):
if (gender == male):
sendPromoiton(shirt)
else
sendPromotion(dress)
Algorithm knows to suggest for gender, but not buying patterns, age, occasion, etc… is it
intelligent?
What?
Tic-Tac-Toe Algorithm
 Let’s pseudo code an algorithm right now
 If you were playing Tic-Tac-Toe, how would you decide to move?
 Algorithm: a step by step process (game strategy) for completing a task (winning)
What?
Tic-Tac-Toe Algorithm
 Check if we have 2 in a row next to an empty space, play and win
 Check if opponent has 2 in a row next to an empty space, block it
 Imagine playing in a space and how opponent would react… repeat
 Try to play in spaces that maximize my connections while minimizing opponent’s
 It’s just tic-tac-toe, it doesn’t matter that much, when in doubt choose randomly and
remember what happens for next time (experiment)
What?
Business Use Case #1:
Segmenting Customers
 Customer Personas
 ‘a semi-fictional representation of your ideal customer based on market
research and real data about your existing customers’
 Allow for targeted marketing messages
 Personalize = higher conversions
 Previous method: manually identify, sort, and maintain separate lists
 Problem: expensive (time and money), prone to human error, not standardized
therefore not improvable
What?
Segmenting Customer Personas
 Challenge: find a more repeatable, scalable process for sorting customers into
distinct user personas
 Type of problem: clustering (grouping)
 Algorithm: K-means clustering
 Why:
 Groups data into distinct clusters
 Doesn’t need to know any labels or additional information (unsupervised)
 Can be used to label data for future categorization
What?
K-Means Clustering: Details
 Goal: Group bunches of points into ‘K’ distinct groups
 Provided Inputs
 Set of Data Points
 Integer value of ‘K’, ie 3 meaning split data points into 3 clusters
 Outputs
 K number of ranges containing all provided data points
 Note this is not same as categorization (unsupervised)
How?
K-Means Clustering: Process
 1) Initialize K cluster points centers, called
‘centroids’ at random locations
 2) For each point, calculate distance to centroids
and assign to closest centroid (smallest
distance)
 3) Update centroid to average position of all data
points in its cluster
 4) Repeat steps 2 and 3 until clusters do not
change from one run to next
 5) Evaluate model: Silhouette Coefficient
How?
K-Means Clustering: Process
How?
 Example of how clusters change per
iteration
 Here the random initial centroid spots
create a ‘green’ cluster that is imprecise,
and a ‘blue’ cluster spread between 2
clusters
 As a result, the blue centroid is ‘pulled’
towards its center towards top middle,
thus taking more out of green and shifting
green to bottom left
K-Means Clustering: Process
How?
Use Case #1- Clustering Personas
Summary
 High Level
 Separating user personas is a situation with a lot of unlabeled data
 KMeans clustering can be used to group data points into K distinct groups
 Advantage is that is relatively easy to implement
 Deeper Dive
 An iterative algorithm which runs many times
 Optimizes centroids at the average point of all the points within their cluster
 Questions?
Business Use Case #2:
Product Recommendations
 Product Recommendations
 Allow for personalized advertising, complementary buys, and upsells
 Maximize each customer’s lifetime value
 Previous method: one-size-fits-all recommendations
 Problem: not personalized, can be operationally difficult
What?
Product Recommendations
 Challenge: generate personalized recommendations for each individual user, not
just broad categories of users
 Type of problem: neighbor distance calculation
 Algorithm: K-Nearest Neighbors (KNN)
 Why:
 Calculates nearest neighbors to any given data point
 Relatively simple to implement with high output quality
 Can incorporate various sources of data: product characteristics or
characteristics of users who also bought, special logic (context)
What?
KNN: Details
 Goal: Find the most similar items to a given data point by mapping out the entire
universe of relevant points
 Provided Inputs
 Specific data point
 Universe of data points
 K – number of neighbors to return
 Method to calculate similarity
 Outputs
 K neighbors closest (most similar) to provided input data point
How?
KNN Cosine Similarity: Side Note
 Side note: Why cosine similarity?
 We must first answer, what are vectors?
 Distance between two points is a function of two
elements:
 Magnitude
 Direction
 Vectors are combinations of magnitudes and direction,
and multi-dimensional vectors can be broken down into
smaller parts (ie x and y)
 Allows us to create a single vector which expresses
multiple different metrics, such as 1) user rating and 2)
price
How?
KNN Cosine Similarity: Side Note
 Side note: Why cosine similarity?
 Multiple ways of measuring similarity between
two items
 Pure distance between two things isn’t always
best measure
 Consider case of direction as positive or
negative ratings
 End distance from points not as important as
similarity in vectors
How?
KNN Cosine Similarity: Process
 1) Clean, wrangle and normalize your data
 2) Pick a point from data set and calculate
distance (cosine similarity) from given point
 3) Repeat for all points in data set
 4) Return K choices with highest similarities
How?
KNN Cosine Similarity: Process
How?
 1) Prepare inputs
 Select columns: style_attributes & mrp
 Clean data and convert into correct numerical types
 Normalise data using the feature scaling and ordinal scaling
techniques
 Store inputs in correct data structure, i.e. dictionary in this case
 2) Define a function to calculate distance between any two points
 3) Write function to iterate distances between primary point to find it’s
closest K neighbors
 4) Return neighbors as suggestions
 Let’s look at the code!
Use Case #2- Product
Recommendations
 High Level
 Using datasets in different segments to make more personalized recommendations to
customers
 Increase basket size and average order value to drive sales and improve customer experience
 Advantage: automate reccommendations to customers on the website/eDM/ads
 Deeper Dive
 A non-parametric, lazy algorithm that returns closest matches given a starting point and
number of desired recommendations
 Uses some type of distance metric to compute distance, and returns closest neighbors
 Questions?
Practical Tips and Tools For ML & AI in
eCommerce
 NLP – it’s complex under the hood, but easy to implement
 Sentiment analysis for reviews: https://www.lexalytics.com/
 Chatbot platform with lots of easy integrations: API.ai
 Python – many powerful libraries to start analyzing your data today
 Scikit-learn, SciPy, StatsModels, PySpark, NLTK and many others
 Cloud services for running recommendation engines in real-time
 Enterprise Cloud Solutions for Deployment (e.g. AWS EMR + Redshift + Elastic
Beanstalk)
Matt R O’Connor
http://Reboot.ai
6289.9447
matt.oconnor217@gmail.com
linkedin.com/in/mattroconnor/
Dhruv Sahi
http://Reboot.ai
5572 8474 dhruvsahi@gmail.com
linkedin.com/in/dhruv-sahi/
Resources
 http://bigdata-madesimple.com/possibly-the-simplest-way-to-explain-k-means-algorithm/
 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4978658/
 https://saravananthirumuruganathan.wordpress.com/2010/05/17/a-detailed-introduction-to-k-nearest-neighbor-knn-algorithm/
 https://www.youtube.com/watch?v=C-JauEnlSlM

More Related Content

What's hot

Tweets Classification using Naive Bayes and SVM
Tweets Classification using Naive Bayes and SVMTweets Classification using Naive Bayes and SVM
Tweets Classification using Naive Bayes and SVMTrilok Sharma
 
Introduction to Machine Learning
Introduction to Machine LearningIntroduction to Machine Learning
Introduction to Machine LearningLior Rokach
 
Internship project presentation_final_upload
Internship project presentation_final_uploadInternship project presentation_final_upload
Internship project presentation_final_uploadSuraj Rathore
 
Automation of IT Ticket Automation using NLP and Deep Learning
Automation of IT Ticket Automation using NLP and Deep LearningAutomation of IT Ticket Automation using NLP and Deep Learning
Automation of IT Ticket Automation using NLP and Deep LearningPranov Mishra
 
Machine learning algorithms and business use cases
Machine learning algorithms and business use casesMachine learning algorithms and business use cases
Machine learning algorithms and business use casesSridhar Ratakonda
 
Three case studies deploying cluster analysis
Three case studies deploying cluster analysisThree case studies deploying cluster analysis
Three case studies deploying cluster analysisGreg Makowski
 
User Payment Prediction in Free-to-Play
User Payment Prediction in Free-to-PlayUser Payment Prediction in Free-to-Play
User Payment Prediction in Free-to-PlayAhmed Hassan
 
Machine Learning Lecture 3 Decision Trees
Machine Learning Lecture 3 Decision TreesMachine Learning Lecture 3 Decision Trees
Machine Learning Lecture 3 Decision Treesananth
 
MachineLearning.ppt
MachineLearning.pptMachineLearning.ppt
MachineLearning.pptbutest
 
Machine Learning: Foundations Course Number 0368403401
Machine Learning: Foundations Course Number 0368403401Machine Learning: Foundations Course Number 0368403401
Machine Learning: Foundations Course Number 0368403401butest
 
Machine learning algorithms
Machine learning algorithmsMachine learning algorithms
Machine learning algorithmsShalitha Suranga
 
Machine Learning 1 - Introduction
Machine Learning 1 - IntroductionMachine Learning 1 - Introduction
Machine Learning 1 - Introductionbutest
 
Dimensionality Reduction
Dimensionality ReductionDimensionality Reduction
Dimensionality ReductionSaad Elbeleidy
 

What's hot (16)

Tweets Classification using Naive Bayes and SVM
Tweets Classification using Naive Bayes and SVMTweets Classification using Naive Bayes and SVM
Tweets Classification using Naive Bayes and SVM
 
Introduction to Machine Learning
Introduction to Machine LearningIntroduction to Machine Learning
Introduction to Machine Learning
 
Internship project presentation_final_upload
Internship project presentation_final_uploadInternship project presentation_final_upload
Internship project presentation_final_upload
 
Automation of IT Ticket Automation using NLP and Deep Learning
Automation of IT Ticket Automation using NLP and Deep LearningAutomation of IT Ticket Automation using NLP and Deep Learning
Automation of IT Ticket Automation using NLP and Deep Learning
 
Data mining
Data miningData mining
Data mining
 
Machine learning algorithms and business use cases
Machine learning algorithms and business use casesMachine learning algorithms and business use cases
Machine learning algorithms and business use cases
 
Three case studies deploying cluster analysis
Three case studies deploying cluster analysisThree case studies deploying cluster analysis
Three case studies deploying cluster analysis
 
User Payment Prediction in Free-to-Play
User Payment Prediction in Free-to-PlayUser Payment Prediction in Free-to-Play
User Payment Prediction in Free-to-Play
 
Machine Learning Lecture 3 Decision Trees
Machine Learning Lecture 3 Decision TreesMachine Learning Lecture 3 Decision Trees
Machine Learning Lecture 3 Decision Trees
 
FLIPKART SAMSUNG
FLIPKART SAMSUNGFLIPKART SAMSUNG
FLIPKART SAMSUNG
 
MachineLearning.ppt
MachineLearning.pptMachineLearning.ppt
MachineLearning.ppt
 
Machine Learning: Foundations Course Number 0368403401
Machine Learning: Foundations Course Number 0368403401Machine Learning: Foundations Course Number 0368403401
Machine Learning: Foundations Course Number 0368403401
 
Machine learning algorithms
Machine learning algorithmsMachine learning algorithms
Machine learning algorithms
 
Decision tree Using c4.5 Algorithm
Decision tree Using c4.5 AlgorithmDecision tree Using c4.5 Algorithm
Decision tree Using c4.5 Algorithm
 
Machine Learning 1 - Introduction
Machine Learning 1 - IntroductionMachine Learning 1 - Introduction
Machine Learning 1 - Introduction
 
Dimensionality Reduction
Dimensionality ReductionDimensionality Reduction
Dimensionality Reduction
 

Similar to Machine Learning in eCommerce: Increasing Sales through Personalization

acmsigtalkshare-121023190142-phpapp01.pptx
acmsigtalkshare-121023190142-phpapp01.pptxacmsigtalkshare-121023190142-phpapp01.pptx
acmsigtalkshare-121023190142-phpapp01.pptxdongchangim30
 
Datamining intro-iep
Datamining intro-iepDatamining intro-iep
Datamining intro-iepaaryarun1999
 
Artificial Intelligence at LinkedIn
Artificial Intelligence at LinkedInArtificial Intelligence at LinkedIn
Artificial Intelligence at LinkedInBill Liu
 
Dive into Machine Learning Event MUGDSC.pptx
Dive into Machine Learning Event MUGDSC.pptxDive into Machine Learning Event MUGDSC.pptx
Dive into Machine Learning Event MUGDSC.pptxRakshaAgrawal21
 
Dive into Machine Learning Event--MUGDSC
Dive into Machine Learning Event--MUGDSCDive into Machine Learning Event--MUGDSC
Dive into Machine Learning Event--MUGDSCRakshaAgrawal21
 
Chapter01.ppt
Chapter01.pptChapter01.ppt
Chapter01.pptbutest
 
The Power of Auto ML and How Does it Work
The Power of Auto ML and How Does it WorkThe Power of Auto ML and How Does it Work
The Power of Auto ML and How Does it WorkIvo Andreev
 
Machine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdfMachine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdfCarlos Paredes
 
Solving churn challenge in Big Data environment - Jelena Pekez
Solving churn challenge in Big Data environment  - Jelena PekezSolving churn challenge in Big Data environment  - Jelena Pekez
Solving churn challenge in Big Data environment - Jelena PekezInstitute of Contemporary Sciences
 
Dwdm ppt for the btech student contain basis
Dwdm ppt for the btech student contain basisDwdm ppt for the btech student contain basis
Dwdm ppt for the btech student contain basisnivatripathy93
 
Scaling & Transforming Stitch Fix's Visibility into What Folks will love
Scaling & Transforming Stitch Fix's Visibility into What Folks will loveScaling & Transforming Stitch Fix's Visibility into What Folks will love
Scaling & Transforming Stitch Fix's Visibility into What Folks will loveJune Andrews
 
A business level introduction to Artificial Intelligence - Louis Dorard @ PAP...
A business level introduction to Artificial Intelligence - Louis Dorard @ PAP...A business level introduction to Artificial Intelligence - Louis Dorard @ PAP...
A business level introduction to Artificial Intelligence - Louis Dorard @ PAP...PAPIs.io
 
Lessons Learned from Building Machine Learning Software at Netflix
Lessons Learned from Building Machine Learning Software at NetflixLessons Learned from Building Machine Learning Software at Netflix
Lessons Learned from Building Machine Learning Software at NetflixJustin Basilico
 
Advances in Exploratory Data Analysis, Visualisation and Quality for Data Cen...
Advances in Exploratory Data Analysis, Visualisation and Quality for Data Cen...Advances in Exploratory Data Analysis, Visualisation and Quality for Data Cen...
Advances in Exploratory Data Analysis, Visualisation and Quality for Data Cen...Hima Patel
 
Tutorial Knowledge Discovery
Tutorial Knowledge DiscoveryTutorial Knowledge Discovery
Tutorial Knowledge DiscoverySSSW
 

Similar to Machine Learning in eCommerce: Increasing Sales through Personalization (20)

LR2. Summary Day 2
LR2. Summary Day 2LR2. Summary Day 2
LR2. Summary Day 2
 
acmsigtalkshare-121023190142-phpapp01.pptx
acmsigtalkshare-121023190142-phpapp01.pptxacmsigtalkshare-121023190142-phpapp01.pptx
acmsigtalkshare-121023190142-phpapp01.pptx
 
Datamining intro-iep
Datamining intro-iepDatamining intro-iep
Datamining intro-iep
 
Artificial Intelligence at LinkedIn
Artificial Intelligence at LinkedInArtificial Intelligence at LinkedIn
Artificial Intelligence at LinkedIn
 
Dive into Machine Learning Event MUGDSC.pptx
Dive into Machine Learning Event MUGDSC.pptxDive into Machine Learning Event MUGDSC.pptx
Dive into Machine Learning Event MUGDSC.pptx
 
Dive into Machine Learning Event--MUGDSC
Dive into Machine Learning Event--MUGDSCDive into Machine Learning Event--MUGDSC
Dive into Machine Learning Event--MUGDSC
 
Chapter01.ppt
Chapter01.pptChapter01.ppt
Chapter01.ppt
 
AI and Deep Learning
AI and Deep Learning AI and Deep Learning
AI and Deep Learning
 
Corporate presentation
Corporate presentationCorporate presentation
Corporate presentation
 
Dwd mdatamining intro-iep
Dwd mdatamining intro-iepDwd mdatamining intro-iep
Dwd mdatamining intro-iep
 
The Power of Auto ML and How Does it Work
The Power of Auto ML and How Does it WorkThe Power of Auto ML and How Does it Work
The Power of Auto ML and How Does it Work
 
Machine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdfMachine_Learning_with_MATLAB_Seminar_Latest.pdf
Machine_Learning_with_MATLAB_Seminar_Latest.pdf
 
Solving churn challenge in Big Data environment - Jelena Pekez
Solving churn challenge in Big Data environment  - Jelena PekezSolving churn challenge in Big Data environment  - Jelena Pekez
Solving churn challenge in Big Data environment - Jelena Pekez
 
Dwdm ppt for the btech student contain basis
Dwdm ppt for the btech student contain basisDwdm ppt for the btech student contain basis
Dwdm ppt for the btech student contain basis
 
07 learning
07 learning07 learning
07 learning
 
Scaling & Transforming Stitch Fix's Visibility into What Folks will love
Scaling & Transforming Stitch Fix's Visibility into What Folks will loveScaling & Transforming Stitch Fix's Visibility into What Folks will love
Scaling & Transforming Stitch Fix's Visibility into What Folks will love
 
A business level introduction to Artificial Intelligence - Louis Dorard @ PAP...
A business level introduction to Artificial Intelligence - Louis Dorard @ PAP...A business level introduction to Artificial Intelligence - Louis Dorard @ PAP...
A business level introduction to Artificial Intelligence - Louis Dorard @ PAP...
 
Lessons Learned from Building Machine Learning Software at Netflix
Lessons Learned from Building Machine Learning Software at NetflixLessons Learned from Building Machine Learning Software at Netflix
Lessons Learned from Building Machine Learning Software at Netflix
 
Advances in Exploratory Data Analysis, Visualisation and Quality for Data Cen...
Advances in Exploratory Data Analysis, Visualisation and Quality for Data Cen...Advances in Exploratory Data Analysis, Visualisation and Quality for Data Cen...
Advances in Exploratory Data Analysis, Visualisation and Quality for Data Cen...
 
Tutorial Knowledge Discovery
Tutorial Knowledge DiscoveryTutorial Knowledge Discovery
Tutorial Knowledge Discovery
 

Recently uploaded

Call Girls In DLf Gurgaon ➥99902@11544 ( Best price)100% Genuine Escort In 24...
Call Girls In DLf Gurgaon ➥99902@11544 ( Best price)100% Genuine Escort In 24...Call Girls In DLf Gurgaon ➥99902@11544 ( Best price)100% Genuine Escort In 24...
Call Girls In DLf Gurgaon ➥99902@11544 ( Best price)100% Genuine Escort In 24...lizamodels9
 
MONA 98765-12871 CALL GIRLS IN LUDHIANA LUDHIANA CALL GIRL
MONA 98765-12871 CALL GIRLS IN LUDHIANA LUDHIANA CALL GIRLMONA 98765-12871 CALL GIRLS IN LUDHIANA LUDHIANA CALL GIRL
MONA 98765-12871 CALL GIRLS IN LUDHIANA LUDHIANA CALL GIRLSeo
 
Ensure the security of your HCL environment by applying the Zero Trust princi...
Ensure the security of your HCL environment by applying the Zero Trust princi...Ensure the security of your HCL environment by applying the Zero Trust princi...
Ensure the security of your HCL environment by applying the Zero Trust princi...Roland Driesen
 
BEST ✨ Call Girls In Indirapuram Ghaziabad ✔️ 9871031762 ✔️ Escorts Service...
BEST ✨ Call Girls In  Indirapuram Ghaziabad  ✔️ 9871031762 ✔️ Escorts Service...BEST ✨ Call Girls In  Indirapuram Ghaziabad  ✔️ 9871031762 ✔️ Escorts Service...
BEST ✨ Call Girls In Indirapuram Ghaziabad ✔️ 9871031762 ✔️ Escorts Service...noida100girls
 
Monte Carlo simulation : Simulation using MCSM
Monte Carlo simulation : Simulation using MCSMMonte Carlo simulation : Simulation using MCSM
Monte Carlo simulation : Simulation using MCSMRavindra Nath Shukla
 
M.C Lodges -- Guest House in Jhang.
M.C Lodges --  Guest House in Jhang.M.C Lodges --  Guest House in Jhang.
M.C Lodges -- Guest House in Jhang.Aaiza Hassan
 
VIP Kolkata Call Girl Howrah 👉 8250192130 Available With Room
VIP Kolkata Call Girl Howrah 👉 8250192130  Available With RoomVIP Kolkata Call Girl Howrah 👉 8250192130  Available With Room
VIP Kolkata Call Girl Howrah 👉 8250192130 Available With Roomdivyansh0kumar0
 
Grateful 7 speech thanking everyone that has helped.pdf
Grateful 7 speech thanking everyone that has helped.pdfGrateful 7 speech thanking everyone that has helped.pdf
Grateful 7 speech thanking everyone that has helped.pdfPaul Menig
 
VIP Call Girls In Saharaganj ( Lucknow ) 🔝 8923113531 🔝 Cash Payment (COD) 👒
VIP Call Girls In Saharaganj ( Lucknow  ) 🔝 8923113531 🔝  Cash Payment (COD) 👒VIP Call Girls In Saharaganj ( Lucknow  ) 🔝 8923113531 🔝  Cash Payment (COD) 👒
VIP Call Girls In Saharaganj ( Lucknow ) 🔝 8923113531 🔝 Cash Payment (COD) 👒anilsa9823
 
Call Girls in Gomti Nagar - 7388211116 - With room Service
Call Girls in Gomti Nagar - 7388211116  - With room ServiceCall Girls in Gomti Nagar - 7388211116  - With room Service
Call Girls in Gomti Nagar - 7388211116 - With room Servicediscovermytutordmt
 
Mondelez State of Snacking and Future Trends 2023
Mondelez State of Snacking and Future Trends 2023Mondelez State of Snacking and Future Trends 2023
Mondelez State of Snacking and Future Trends 2023Neil Kimberley
 
Keppel Ltd. 1Q 2024 Business Update Presentation Slides
Keppel Ltd. 1Q 2024 Business Update  Presentation SlidesKeppel Ltd. 1Q 2024 Business Update  Presentation Slides
Keppel Ltd. 1Q 2024 Business Update Presentation SlidesKeppelCorporation
 
Call Girls Navi Mumbai Just Call 9907093804 Top Class Call Girl Service Avail...
Call Girls Navi Mumbai Just Call 9907093804 Top Class Call Girl Service Avail...Call Girls Navi Mumbai Just Call 9907093804 Top Class Call Girl Service Avail...
Call Girls Navi Mumbai Just Call 9907093804 Top Class Call Girl Service Avail...Dipal Arora
 
A DAY IN THE LIFE OF A SALESMAN / WOMAN
A DAY IN THE LIFE OF A  SALESMAN / WOMANA DAY IN THE LIFE OF A  SALESMAN / WOMAN
A DAY IN THE LIFE OF A SALESMAN / WOMANIlamathiKannappan
 
Tech Startup Growth Hacking 101 - Basics on Growth Marketing
Tech Startup Growth Hacking 101  - Basics on Growth MarketingTech Startup Growth Hacking 101  - Basics on Growth Marketing
Tech Startup Growth Hacking 101 - Basics on Growth MarketingShawn Pang
 
KYC-Verified Accounts: Helping Companies Handle Challenging Regulatory Enviro...
KYC-Verified Accounts: Helping Companies Handle Challenging Regulatory Enviro...KYC-Verified Accounts: Helping Companies Handle Challenging Regulatory Enviro...
KYC-Verified Accounts: Helping Companies Handle Challenging Regulatory Enviro...Any kyc Account
 
Event mailer assignment progress report .pdf
Event mailer assignment progress report .pdfEvent mailer assignment progress report .pdf
Event mailer assignment progress report .pdftbatkhuu1
 
HONOR Veterans Event Keynote by Michael Hawkins
HONOR Veterans Event Keynote by Michael HawkinsHONOR Veterans Event Keynote by Michael Hawkins
HONOR Veterans Event Keynote by Michael HawkinsMichael W. Hawkins
 
Progress Report - Oracle Database Analyst Summit
Progress  Report - Oracle Database Analyst SummitProgress  Report - Oracle Database Analyst Summit
Progress Report - Oracle Database Analyst SummitHolger Mueller
 
Creating Low-Code Loan Applications using the Trisotech Mortgage Feature Set
Creating Low-Code Loan Applications using the Trisotech Mortgage Feature SetCreating Low-Code Loan Applications using the Trisotech Mortgage Feature Set
Creating Low-Code Loan Applications using the Trisotech Mortgage Feature SetDenis Gagné
 

Recently uploaded (20)

Call Girls In DLf Gurgaon ➥99902@11544 ( Best price)100% Genuine Escort In 24...
Call Girls In DLf Gurgaon ➥99902@11544 ( Best price)100% Genuine Escort In 24...Call Girls In DLf Gurgaon ➥99902@11544 ( Best price)100% Genuine Escort In 24...
Call Girls In DLf Gurgaon ➥99902@11544 ( Best price)100% Genuine Escort In 24...
 
MONA 98765-12871 CALL GIRLS IN LUDHIANA LUDHIANA CALL GIRL
MONA 98765-12871 CALL GIRLS IN LUDHIANA LUDHIANA CALL GIRLMONA 98765-12871 CALL GIRLS IN LUDHIANA LUDHIANA CALL GIRL
MONA 98765-12871 CALL GIRLS IN LUDHIANA LUDHIANA CALL GIRL
 
Ensure the security of your HCL environment by applying the Zero Trust princi...
Ensure the security of your HCL environment by applying the Zero Trust princi...Ensure the security of your HCL environment by applying the Zero Trust princi...
Ensure the security of your HCL environment by applying the Zero Trust princi...
 
BEST ✨ Call Girls In Indirapuram Ghaziabad ✔️ 9871031762 ✔️ Escorts Service...
BEST ✨ Call Girls In  Indirapuram Ghaziabad  ✔️ 9871031762 ✔️ Escorts Service...BEST ✨ Call Girls In  Indirapuram Ghaziabad  ✔️ 9871031762 ✔️ Escorts Service...
BEST ✨ Call Girls In Indirapuram Ghaziabad ✔️ 9871031762 ✔️ Escorts Service...
 
Monte Carlo simulation : Simulation using MCSM
Monte Carlo simulation : Simulation using MCSMMonte Carlo simulation : Simulation using MCSM
Monte Carlo simulation : Simulation using MCSM
 
M.C Lodges -- Guest House in Jhang.
M.C Lodges --  Guest House in Jhang.M.C Lodges --  Guest House in Jhang.
M.C Lodges -- Guest House in Jhang.
 
VIP Kolkata Call Girl Howrah 👉 8250192130 Available With Room
VIP Kolkata Call Girl Howrah 👉 8250192130  Available With RoomVIP Kolkata Call Girl Howrah 👉 8250192130  Available With Room
VIP Kolkata Call Girl Howrah 👉 8250192130 Available With Room
 
Grateful 7 speech thanking everyone that has helped.pdf
Grateful 7 speech thanking everyone that has helped.pdfGrateful 7 speech thanking everyone that has helped.pdf
Grateful 7 speech thanking everyone that has helped.pdf
 
VIP Call Girls In Saharaganj ( Lucknow ) 🔝 8923113531 🔝 Cash Payment (COD) 👒
VIP Call Girls In Saharaganj ( Lucknow  ) 🔝 8923113531 🔝  Cash Payment (COD) 👒VIP Call Girls In Saharaganj ( Lucknow  ) 🔝 8923113531 🔝  Cash Payment (COD) 👒
VIP Call Girls In Saharaganj ( Lucknow ) 🔝 8923113531 🔝 Cash Payment (COD) 👒
 
Call Girls in Gomti Nagar - 7388211116 - With room Service
Call Girls in Gomti Nagar - 7388211116  - With room ServiceCall Girls in Gomti Nagar - 7388211116  - With room Service
Call Girls in Gomti Nagar - 7388211116 - With room Service
 
Mondelez State of Snacking and Future Trends 2023
Mondelez State of Snacking and Future Trends 2023Mondelez State of Snacking and Future Trends 2023
Mondelez State of Snacking and Future Trends 2023
 
Keppel Ltd. 1Q 2024 Business Update Presentation Slides
Keppel Ltd. 1Q 2024 Business Update  Presentation SlidesKeppel Ltd. 1Q 2024 Business Update  Presentation Slides
Keppel Ltd. 1Q 2024 Business Update Presentation Slides
 
Call Girls Navi Mumbai Just Call 9907093804 Top Class Call Girl Service Avail...
Call Girls Navi Mumbai Just Call 9907093804 Top Class Call Girl Service Avail...Call Girls Navi Mumbai Just Call 9907093804 Top Class Call Girl Service Avail...
Call Girls Navi Mumbai Just Call 9907093804 Top Class Call Girl Service Avail...
 
A DAY IN THE LIFE OF A SALESMAN / WOMAN
A DAY IN THE LIFE OF A  SALESMAN / WOMANA DAY IN THE LIFE OF A  SALESMAN / WOMAN
A DAY IN THE LIFE OF A SALESMAN / WOMAN
 
Tech Startup Growth Hacking 101 - Basics on Growth Marketing
Tech Startup Growth Hacking 101  - Basics on Growth MarketingTech Startup Growth Hacking 101  - Basics on Growth Marketing
Tech Startup Growth Hacking 101 - Basics on Growth Marketing
 
KYC-Verified Accounts: Helping Companies Handle Challenging Regulatory Enviro...
KYC-Verified Accounts: Helping Companies Handle Challenging Regulatory Enviro...KYC-Verified Accounts: Helping Companies Handle Challenging Regulatory Enviro...
KYC-Verified Accounts: Helping Companies Handle Challenging Regulatory Enviro...
 
Event mailer assignment progress report .pdf
Event mailer assignment progress report .pdfEvent mailer assignment progress report .pdf
Event mailer assignment progress report .pdf
 
HONOR Veterans Event Keynote by Michael Hawkins
HONOR Veterans Event Keynote by Michael HawkinsHONOR Veterans Event Keynote by Michael Hawkins
HONOR Veterans Event Keynote by Michael Hawkins
 
Progress Report - Oracle Database Analyst Summit
Progress  Report - Oracle Database Analyst SummitProgress  Report - Oracle Database Analyst Summit
Progress Report - Oracle Database Analyst Summit
 
Creating Low-Code Loan Applications using the Trisotech Mortgage Feature Set
Creating Low-Code Loan Applications using the Trisotech Mortgage Feature SetCreating Low-Code Loan Applications using the Trisotech Mortgage Feature Set
Creating Low-Code Loan Applications using the Trisotech Mortgage Feature Set
 

Machine Learning in eCommerce: Increasing Sales through Personalization

  • 1.
  • 3.
  • 4. Free PowerPoint Templates HotDesk Fixed DeskPrivate Offices Event Space rental Kennedy Town
  • 5. Free PowerPoint Templates Wong Chuk Hang Photo studio Hot desk Runway Fixed Desk Private Offices Event Space rental
  • 6.
  • 7.
  • 10. Machine Learning in eCommerce USING MACHINE LEARNING TO INCREASE CONVERSIONS AND SALES REBOOT.AI
  • 11. Agenda  Who: Who are we?  Why: Why value does machine learning add in eCommerce?  What: What algorithms are used in eCommerce?  Introduction to algorithms  Business Use Cases : User Personas, Product Recommendations  Which algorithm is used for this specific case?  How: How does this algorithm actually work?  High Level Description  Example Implementation  Q&A
  • 12. About Reboot.ai  Matt O’Connor  BBA Finance  Previous: Lead Trader Algorithmic Desk - Macro Hedge Fund  Current: Full stack developer and professional Scrum Master (PSM I)  Avid futurist –social ramifications of AI & blockchain  Dhruv Sahi  BA Mathematics and Economics  Previous: Data Science Chief – Smart Cities Startup  Current: Business Intelligence Analyst – eCommerce - Grana  AI, IoT, and smart cities enthusiast  Reboot.ai  Hong Kong’s only dedicated machine learning and AI training provider  Part time evening courses for beginners and advanced  Curriculums developed in partnership with local data companies  Use ML & AI in our classrooms to improve teaching and personalize learning Who?
  • 13. Why Machine Learning?  1) Computers much faster than humans  Even complex or infinite solution problems have practical ‘solutions’ and optimizations  Ex. Google maps vs human intuition  2) Logic is replicable and scalable  Consistency of results not humanly possible  Conducive to experimentation and A/B testing can limit variables at play  3) Can incorporate elements of ‘learning’ from results  Can ‘teach itself’ and improve  Can identify insights that are not intuitive or sometimes invisible to humans Why?
  • 14. Headline Use Cases  Recommendation Engines: How Amazon and Netflix Are Winning the Personalization Battle and optimizing revenues  75% of all content on Netflix is viewed through their recommendation engine  35% of Amazon’s revenues are the product of their recommendation engine  Machine Learning Generates Clickbait Headlines That Will SHOCK You  Predict Sentiment From Movie Reviews Using Deep Learning  Can Chatbots Help Reduce Customer Service Costs by 30%? Why?
  • 15. Separating Two Customer Groups… Why?
  • 16. …Is Easy for a machine Why?
  • 18. Summary: Why?  Benefits  Can be faster and cheaper than human alternative  Can be employed in a wide variety of real world conditions even with limited/flawed data  Can improve, learn, and identify trends humans would have trouble identifying  Weaknesses  Very difficult to create intelligence good in multiple unrelated contexts  No instincts, ‘genetic knowledge’ or ‘intuition’  Mistrusted and misunderstood  Questions? Why?
  • 19. What is an Algorithm  An algorithm is a step by step process for completing a task.  Everyday examples: recipes, ‘habits’, traditions, traffic laws  Example in code emailCustomer(gender): if (gender == male): sendPromoiton(shirt) else sendPromotion(dress) Algorithm knows to suggest for gender, but not buying patterns, age, occasion, etc… is it intelligent? What?
  • 20. Tic-Tac-Toe Algorithm  Let’s pseudo code an algorithm right now  If you were playing Tic-Tac-Toe, how would you decide to move?  Algorithm: a step by step process (game strategy) for completing a task (winning) What?
  • 21. Tic-Tac-Toe Algorithm  Check if we have 2 in a row next to an empty space, play and win  Check if opponent has 2 in a row next to an empty space, block it  Imagine playing in a space and how opponent would react… repeat  Try to play in spaces that maximize my connections while minimizing opponent’s  It’s just tic-tac-toe, it doesn’t matter that much, when in doubt choose randomly and remember what happens for next time (experiment) What?
  • 22. Business Use Case #1: Segmenting Customers  Customer Personas  ‘a semi-fictional representation of your ideal customer based on market research and real data about your existing customers’  Allow for targeted marketing messages  Personalize = higher conversions  Previous method: manually identify, sort, and maintain separate lists  Problem: expensive (time and money), prone to human error, not standardized therefore not improvable What?
  • 23. Segmenting Customer Personas  Challenge: find a more repeatable, scalable process for sorting customers into distinct user personas  Type of problem: clustering (grouping)  Algorithm: K-means clustering  Why:  Groups data into distinct clusters  Doesn’t need to know any labels or additional information (unsupervised)  Can be used to label data for future categorization What?
  • 24. K-Means Clustering: Details  Goal: Group bunches of points into ‘K’ distinct groups  Provided Inputs  Set of Data Points  Integer value of ‘K’, ie 3 meaning split data points into 3 clusters  Outputs  K number of ranges containing all provided data points  Note this is not same as categorization (unsupervised) How?
  • 25. K-Means Clustering: Process  1) Initialize K cluster points centers, called ‘centroids’ at random locations  2) For each point, calculate distance to centroids and assign to closest centroid (smallest distance)  3) Update centroid to average position of all data points in its cluster  4) Repeat steps 2 and 3 until clusters do not change from one run to next  5) Evaluate model: Silhouette Coefficient How?
  • 26. K-Means Clustering: Process How?  Example of how clusters change per iteration  Here the random initial centroid spots create a ‘green’ cluster that is imprecise, and a ‘blue’ cluster spread between 2 clusters  As a result, the blue centroid is ‘pulled’ towards its center towards top middle, thus taking more out of green and shifting green to bottom left
  • 28. Use Case #1- Clustering Personas Summary  High Level  Separating user personas is a situation with a lot of unlabeled data  KMeans clustering can be used to group data points into K distinct groups  Advantage is that is relatively easy to implement  Deeper Dive  An iterative algorithm which runs many times  Optimizes centroids at the average point of all the points within their cluster  Questions?
  • 29. Business Use Case #2: Product Recommendations  Product Recommendations  Allow for personalized advertising, complementary buys, and upsells  Maximize each customer’s lifetime value  Previous method: one-size-fits-all recommendations  Problem: not personalized, can be operationally difficult What?
  • 30. Product Recommendations  Challenge: generate personalized recommendations for each individual user, not just broad categories of users  Type of problem: neighbor distance calculation  Algorithm: K-Nearest Neighbors (KNN)  Why:  Calculates nearest neighbors to any given data point  Relatively simple to implement with high output quality  Can incorporate various sources of data: product characteristics or characteristics of users who also bought, special logic (context) What?
  • 31. KNN: Details  Goal: Find the most similar items to a given data point by mapping out the entire universe of relevant points  Provided Inputs  Specific data point  Universe of data points  K – number of neighbors to return  Method to calculate similarity  Outputs  K neighbors closest (most similar) to provided input data point How?
  • 32. KNN Cosine Similarity: Side Note  Side note: Why cosine similarity?  We must first answer, what are vectors?  Distance between two points is a function of two elements:  Magnitude  Direction  Vectors are combinations of magnitudes and direction, and multi-dimensional vectors can be broken down into smaller parts (ie x and y)  Allows us to create a single vector which expresses multiple different metrics, such as 1) user rating and 2) price How?
  • 33. KNN Cosine Similarity: Side Note  Side note: Why cosine similarity?  Multiple ways of measuring similarity between two items  Pure distance between two things isn’t always best measure  Consider case of direction as positive or negative ratings  End distance from points not as important as similarity in vectors How?
  • 34. KNN Cosine Similarity: Process  1) Clean, wrangle and normalize your data  2) Pick a point from data set and calculate distance (cosine similarity) from given point  3) Repeat for all points in data set  4) Return K choices with highest similarities How?
  • 35. KNN Cosine Similarity: Process How?  1) Prepare inputs  Select columns: style_attributes & mrp  Clean data and convert into correct numerical types  Normalise data using the feature scaling and ordinal scaling techniques  Store inputs in correct data structure, i.e. dictionary in this case  2) Define a function to calculate distance between any two points  3) Write function to iterate distances between primary point to find it’s closest K neighbors  4) Return neighbors as suggestions  Let’s look at the code!
  • 36. Use Case #2- Product Recommendations  High Level  Using datasets in different segments to make more personalized recommendations to customers  Increase basket size and average order value to drive sales and improve customer experience  Advantage: automate reccommendations to customers on the website/eDM/ads  Deeper Dive  A non-parametric, lazy algorithm that returns closest matches given a starting point and number of desired recommendations  Uses some type of distance metric to compute distance, and returns closest neighbors  Questions?
  • 37. Practical Tips and Tools For ML & AI in eCommerce  NLP – it’s complex under the hood, but easy to implement  Sentiment analysis for reviews: https://www.lexalytics.com/  Chatbot platform with lots of easy integrations: API.ai  Python – many powerful libraries to start analyzing your data today  Scikit-learn, SciPy, StatsModels, PySpark, NLTK and many others  Cloud services for running recommendation engines in real-time  Enterprise Cloud Solutions for Deployment (e.g. AWS EMR + Redshift + Elastic Beanstalk)
  • 38. Matt R O’Connor http://Reboot.ai 6289.9447 matt.oconnor217@gmail.com linkedin.com/in/mattroconnor/ Dhruv Sahi http://Reboot.ai 5572 8474 dhruvsahi@gmail.com linkedin.com/in/dhruv-sahi/
  • 39. Resources  http://bigdata-madesimple.com/possibly-the-simplest-way-to-explain-k-means-algorithm/  https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4978658/  https://saravananthirumuruganathan.wordpress.com/2010/05/17/a-detailed-introduction-to-k-nearest-neighbor-knn-algorithm/  https://www.youtube.com/watch?v=C-JauEnlSlM

Editor's Notes

  1. Campfire KT: Digital and tech environment focus. We offer solution for: every company/team size. every industry Need to target the need of your prospect. Listen to him/her and propose accordingly.
  2. Fashion, Design and Creative spirit.
  3. Campfire is planning to become the new ecosystem for (net)work. Inspired work environment Networking Value added Service Weekly events ----------------------------------------------------- 1) Campfire Secret Island Party – a two day outdoor event 2) Campfire Networking Thurdays – Pitch night event. 3) Campfire Waffle Wednesday – Monthly networking Breakfast event that are host by inspiring guest speakers. 4) Campfire WCH Grande Oppening – Fashion Show for showcasing HK’s talents in the fashion industries. 5) Campfire Yoga Classes – Bi-monthly classes