Predicting Current User Intent with Contextual Markov Models

Predicting Current User Intentwith Contextual Markov Models

Julia Kiseleva, Hoang Thanh Lam, Mykola Pechenizkiy (TU/e)Toon Calders (ULB)

DDDM@ICDM2013, Dallas, TX, USA

CAPA project: http://www.win.tue.nl/~mpechen/projects/capa/

7 December 2013

Outline• What is predictive Web analytics• Context-Aware Predictive Analytics framework• User intent modeling• Contextual Markov Models• Case study, experimental results• Conclusions and further ongoing work

DDDM@ICDM2013Dec 7, 2013

2Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Understanding user needs

DDDM@ICDM2013Dec 7, 2013

3Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Let’s give it a try…

DDDM@ICDM2013Dec 7, 2013

4Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

User Intent Modeling: What?• Next action prediction

– Click prediction in display advertising– Drop out prediction– Trail prediction

• Information need prediction: – Navigational vs. explorative vs. purchase– Open acronym based on context

• Type of product wanted – Personalization based on context

DDDM@ICDM2013Dec 7, 2013

5Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

User Intent Modeling: Why?• To understand users and website usage

– redesign website, redirect flows, – diversified search, recommendations

• To better use budget (pageviews)– what (type of) ads to serve? – brand awareness CPM, or convergence CPC

• To manipulate user – worth giving a promotion?– personalize with intent of converging to a desired

action– personalized suggestions based on user context

DDDM@ICDM2013Dec 7, 2013

6Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

User Intent Modeling: How?

Model L

population(source)

Historicaldata

labels

label?

1. training

2.

2. application

X

y

X'

y'

Training:

y = L (X)

Application:use Lfor an unseen data

y' = L (X')

labels

Testingdata

DDDM@ICDM2013Dec 7, 2013

7Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Context in IR & RecSys

• User Context– Preferences, usage history, profiles

• Document/Product Context– Meta-data, content features

• Task Context– Current activity, location etc.

• Social Context– Leveraging the social graph

DDDM@ICDM2013Dec 7, 2013

8Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Context in Diagnostics Not predictive alone but a subset of features with the contextual attribute(s) becomes (much) more predictive

Time of the daycontext

no context

DDDM@ICDM2013Dec 7, 2013

9Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Context in Marketing P(Purchase|gender=“male”)=P(Purchase|gender=“female”)ModelMale~f(relevance); ModelFemale~f(perceived value)

gendercontext

no context

Male

Female

buy

buy

relevance

relevance

buy

don’t

don’t

don’t

gender

DDDM@ICDM2013Dec 7, 2013

10Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Environment/Context

Model L

population

Training:

??

Application:

y' = Lj (X')Lj <= G(X',E)

X'

y'

Historicaldata

labels

X

y

label?

Context-Awareness as Meta-learning

labels

Testdata

DDDM@ICDM2013Dec 7, 2013

11Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Learning Classifiers & Context

DDDM@ICDM2013Dec 7, 2013

12Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Research Questions• How to define the context (form and maintain contextual categories) in web analytics?

• How to connect context with the prediction process in predictive web analytics?

• How to integrate change detection mechanisms into the prediction process in web analytics?

• How to ensure integration and feedback mechanisms between change detection and context awareness mechanisms?

• What should a reference architecture allowing to plug in new context aware prediction techniques for a collection of web analytics tasks look like?

DDDM@ICDM2013Dec 7, 2013

14Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

IEEE CBMS 2010Perth, Australia

Handling Concept Drift in Medical Applications: Importance, Challenges and Solutions© M. Pechenizkiy and I. Zliobaite

15

• Context-aware ranking of search results

• Drop-out prediction/prevention

• Next action prediction

Mastersportal.eu - Homepage

Quick Search

Banner Click

Universities in the spotlight

Mastersportal.eu - Search

Refine Search

Click on Program is Search Result

Click on University

Click on Country

User Navigation Graph

Motivation for Contextual Markov Models

Useful Contexts: E[M] < pc1*E[Mc1] + pc2*E[Mc2]Why should it help?

Explicit contexts (user location) Implicit contexts (inferred from clickstream)

Implicit Context

Discover clusters in the graph using community detection algorithm

c1 = Novice users

c1 = Experienced

usersC = user type

DDDM@ICDM2013Dec 7, 2013

20Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Dataset

DateSource of information

May 2012Mastersportal.eu

#sessions 350.618#requests 1.775.711

DDDM@ICDM2013Dec 7, 2013

21Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Publicly available at:http://www.win.tue.nl/~mpechen/projects/capa

Accuracy Results

DDDM@ICDM2013Dec 7, 2013

22Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

user location

user type

Global vs. explicit vs. implicit vs. random contexts

DDDM@ICDM2013Dec 7, 2013

23Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Conclusions

• We formulated context discovery as optimization problem

• Our approach can be used to identify useful contexts

• Experiments on a real dataset provide empirical evidence that contextual Markov Models are more accurate than global models

• Further (ongoing) work– Temporal context discovery (TempWeb@WWW’2013)– Multidimensional vertical and horizontal clustering on

the user navigation graphDDDM@ICDM2013Dec 7, 2013

24Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Change of Intent as Context Switch

Timeline

Search Refine Search PaymentClick Product

View Search Click

Context ``Find information”

Context ``Buy product”

What is next?Change of intent?

DDDM@ICDM2013Dec 7, 2013

25Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

User next action prediction

Search Refine Search PaymentClick Product

View Click ?

• What the context is attached to?o Single action?o Session/trail? (user)o A group of sessions (space/time)

• Pattern-mining based approach

Collaboration is welcome!DDDM@ICDM2013Dec 7, 2013

26Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Designing Context-awareness

Predictive model(s) PredictionsTraining data

Context-aware Adaptation

Instance set selectionFeature set selectionFeature set expansion Model selection/weighting

Model adjustment Output correction

if (context == “spring”) select instances(“spring”)

if (context == “spring”) select models (“spring”)

if (context == “spring”) score += 0.1*score

DDDM@ICDM2013Dec 7, 2013

27Predicting Current User Intent with Contextual Markov ModelsMykola Pechenizkiy, Eindhoven University of Technology

Designing Context-awareness

Definitions/properties/

utilities

[Un] [Semi]Super

visedmethods

How to define

context

Context mining: how to discover context

Instance set selectionFeature set selectionFeature set expansion Model selection/weighting

Model adjustment Output correction

Contextual featuresContextual categories

Features not predictive alone, but increasing predictive power of other featuresDescriptors explaining a significant group of instances having some distinct behaviour

Subgroup discoveryAntiLDAUplift modelingActionable attributes

Horizontal Partitioning

Users from Europe

Users from South America

Session 1 Search Refine Search Click on Banner

Product View Payment

Session 3 Product View

Payment

Session 3 Search Refine Search Refine Search Click on Banner

Session 4 Search Refine Search Click on Banner

Product View Payment

Session 5 Product View

Click on Banner Search

Horizontal Partitioning

Two types of behavior:Ready to buy – (Product View, Payment)Just browsing – (Search, Refine Search, Click on

Banner) Session 1 Search Refine

SearchClick on Banner

Product View

Payment

Session 2 Product View

Payment

Session 3 Search Refine Search

Refine Search

Click on Banner

Session 4 Search Refine Search

Click on Banner

Product View

Payment

Session 5 Product View

Click on Banner

Search

Vertical Partitioning

Session 1 Search Refine Search

Click on Banner

Product View

Payment

Session 2 Product View

Payment

Session 3 Search Refine Search

Refine Search

Click on Banner

Session 4 Search Refine Search

Click on Banner

Product View

Payment

Session 5 Product View

Click on Banner

Search

Two types of behavior:Ready to buy – (Product View, Payment)Just browsing – (Search, Refine Search, Click on

Banner)

Vertical Partitioning