Slide 1

Slide 1 text

Justin Cheng, Caroline Lo, Jure Leskovec / Stanford University + Pinterest Predicting Intent Using Activity Logs How Goal Specificity and Temporal Range Affect User Behavior

Slide 2

Slide 2 text

Ajzen (1985) Intent precedes and predicts any future behavior

Slide 3

Slide 3 text

Intent directs how people use systems Intent precedes and predicts any future behavior

Slide 4

Slide 4 text

Intent directs how people use systems Intent precedes and predicts any future behavior

Slide 5

Slide 5 text

Intent prediction enables adaptive user interfaces Intent directs how people use systems Intent precedes and predicts any future behavior Task-specific UI Recommendations

Slide 6

Slide 6 text

But how do we infer intent?

Slide 7

Slide 7 text

Challenge #1: can only observe user behavior But how do we infer intent?

Slide 8

Slide 8 text

Challenge #2: intent can vary significantly Challenge #1: can only observe user behavior But how do we infer intent?

Slide 9

Slide 9 text

Intent varies greatly by individual… (From our preliminary survey of Pinterest users using Mechanical Turk) Bored, just looking around Gardening ideas Searching for car images Finding interesting optical illusions Make a monster truck cake Need a quote of the week Wedding bouquet ideas Recipes for dinner tonight Looking for good food Cupcake recipes for son’s birthday Look up a saved pin To waste time Stock images for my website Fall shoe inspiration Anxiety tips Fitness help

Slide 10

Slide 10 text

…and even for the same individual. (From our preliminary survey of Pinterest users using Mechanical Turk) “I usually go on Pinterest to search for ideas for decorating, furniture, parties, and I also like looking at future stuff like weddings. I sometimes look at travel photos too because it helps me experience things vicariously.”

Slide 11

Slide 11 text

Categorizing intent in search and shopping Prior work

Slide 12

Slide 12 text

Broder (2002) Search is navigational, informational, transactional. Categorizing intent in search and shopping Prior work Looking for a website, for information, or to buy something.

Slide 13

Slide 13 text

Chiou & Ting (2011); Novak, et al. (2003) Shopping is either goal-oriented or experiential. Search is navigational, informational, transactional. Categorizing intent in search and shopping Prior work Buying something specific, or just browsing.

Slide 14

Slide 14 text

How to organize? Bored, just looking around Gardening ideas Searching for car images Finding interesting optical illusions Make a monster truck cake Need a quote of the week Wedding bouquet ideas Recipes for dinner tonight Looking for good food Cupcake recipes for son’s birthday Look up a saved pin To waste time Stock images for my website Fall shoe inspiration Anxiety tips Fitness help

Slide 15

Slide 15 text

Bored, just looking around Gardening ideas Searching for car images Finding interesting optical illusions Make a monster truck cake Need a quote of the week Wedding bouquet ideas Recipes for dinner tonight Looking for good food Cupcake recipes for son’s birthday Look up a saved pin To waste time Stock images for my website Fall shoe inspiration Anxiety tips Fitness help First attempt: goal-oriented vs. experiential?

Slide 16

Slide 16 text

Finding interesting optical illusions First attempt: goal-oriented vs. experiential? Bored, just looking around Gardening ideas Searching for car images Make a monster truck cake Need a quote of the week Wedding bouquet ideas Recipes for dinner tonight Looking for good food Cupcake recipes for son’s birthday Look up a saved pin To waste time Stock images for my website Fall shoe inspiration Anxiety tips Fitness help

Slide 17

Slide 17 text

Finding interesting optical illusions First attempt: goal-oriented vs. experiential? Bored, just looking around Gardening ideas Searching for car images Make a monster truck cake Need a quote of the week Wedding bouquet ideas Recipes for dinner tonight Looking for good food Cupcake recipes for son’s birthday Look up a saved pin To waste time Stock images for my website Fall shoe inspiration Anxiety tips Fitness help

Slide 18

Slide 18 text

Is there a better way to quantify intent?

Slide 19

Slide 19 text

Is there a better way to quantify intent? and predict

Slide 20

Slide 20 text

A generalizable framework for intent prediction This work

Slide 21

Slide 21 text

Large-scale survey + behavioral analysis A generalizable framework for intent prediction This work

Slide 22

Slide 22 text

Identifies two key dimensions of intent Large-scale survey + behavioral analysis A generalizable framework for intent prediction This work Goal specificity Temporal range

Slide 23

Slide 23 text

How does intent affect behavior? Can we predict intent (quickly)? 1 2 3 What is intent?

Slide 24

Slide 24 text

How does intent affect behavior? Can we predict intent (quickly)? 1 2 3 What is intent?

Slide 25

Slide 25 text

How do we define intent?

Slide 26

Slide 26 text

Using two key dimensions of goal-setting: How do we define intent?

Slide 27

Slide 27 text

Hollenbeck & Klein (1987); Locke (1968) Goal specificity: how well-defined is a goal? Using two key dimensions of goal-setting: How do we define intent? Killing time Monster truck cake recipes Gardening ideas Less specific More specific

Slide 28

Slide 28 text

Austin & Vancouver (1996); Frese & Zapf (1994) Temporal range: when will a goal be achieved? Goal specificity: how well-defined is a goal? Using two key dimensions of goal-setting: How do we define intent? Recipe for
 dinner tonight Wedding favors Quote of
 the week Short-term Long-term

Slide 29

Slide 29 text

Goal specificity and temporal range Bored, just looking around Gardening ideas Searching for car images Finding interesting optical illusions Make a monster truck cake Need a quote of the week Wedding bouquet ideas Recipes for dinner tonight Looking for good food Cupcake recipes for son’s birthday Look up a saved pin To waste time Stock images for my website Fall shoe inspiration Anxiety tips Fitness help

Slide 30

Slide 30 text

Goal specificity and temporal range Longer-term More specific Less specific Shorter-term Bored, just looking around Gardening ideas Searching for car images Finding interesting optical illusions Make a monster truck cake Need a quote of the week Wedding bouquet ideas Recipes for dinner tonight Looking for good food Cupcake recipes for stepson’s birthday Look up a saved pin To waste time Stock images for my website Fall shoe inspiration Anxiety tips Fitness help

Slide 31

Slide 31 text

Large-scale survey + behavioral analysis Method

Slide 32

Slide 32 text

(Also in the paper: specific motivations and topics of interest) Users surveyed on goal specificity, temporal range. Large-scale survey + behavioral analysis Method “Are you visiting with a goal in mind?”

Slide 33

Slide 33 text

(Also in the paper: specific motivations and topics of interest) Users surveyed on goal specificity, temporal range. Large-scale survey + behavioral analysis Method “When are you planning to act on it (if applicable)?”

Slide 34

Slide 34 text

We then analyzed their activity in first ten minutes. Users surveyed on goal specificity, temporal range. Large-scale survey + behavioral analysis Method

Slide 35

Slide 35 text

850k interactions across ~6k users Data views, clicks, searches

Slide 36

Slide 36 text

How does goal specificity vary among users? % Users 0% 5% 10% 15% 20% 25% 30% 35% Goal Specificity (1 = Not Specific, 7 = Very Specific) 1 2 3 4 5 6 7

Slide 37

Slide 37 text

Goal specificity is bimodal % Users 0% 5% 10% 15% 20% 25% 30% 35% Goal Specificity (1 = Not Specific, 7 = Very Specific) 1 2 3 4 5 6 7

Slide 38

Slide 38 text

Goal specificity is bimodal % Users 0% 5% 10% 15% 20% 25% 30% 35% Goal Specificity (1 = Not Specific, 7 = Very Specific) 1 2 3 4 5 6 7 Goal-nonspecific Goal-specific

Slide 39

Slide 39 text

How does temporal range vary among users? % Users 0% 5% 10% 15% 20% 25% 30% 35% Short-Term Mid-Term Long-Term Unsure/NA (0-2 days) (3-7 days) (>7 days)

Slide 40

Slide 40 text

Temporal range varies significantly % Users 0% 5% 10% 15% 20% 25% 30% 35% Short-Term Mid-Term Long-Term Unsure/NA (0-2 days) (3-7 days) (>7 days)

Slide 41

Slide 41 text

Goal specificity correlates with temporal range % Users 0% 10% 20% 30% 40% 50% 60% Short-Term Mid-Term Long-Term Unsure/NA Goal-Specific Goal-Nonspecific (0-2 days) (3-7 days) (>7 days)

Slide 42

Slide 42 text

How does intent affect behavior? Can we predict intent (quickly)? 1 2 3 What is intent?

Slide 43

Slide 43 text

How does intent influence search?

Slide 44

Slide 44 text

(p < 10-3, Cohen’s d = 0.42) Goal-specific users search more, How does intent influence search? Goal-specific Goal-nonspecific # searches 0.4 1.1

Slide 45

Slide 45 text

(p < 10-3, d = 0.24) issue more complex queries, Goal-specific users search more, How does intent influence search? Goal-specific Goal-nonspecific # words per search query 2.7 3.0

Slide 46

Slide 46 text

(p < 10-3, d = 0.42) and start searching more quickly. issue more complex queries, Goal-specific users search more, How does intent influence search? Goal-specific Goal-nonspecific Time to first search (minutes) 3.0 1.9

Slide 47

Slide 47 text

How does intent affect browsing? Do goal-specific users browse quickly? Or in detail?

Slide 48

Slide 48 text

How does intent affect browsing? Are users with short-term goals in a hurry?

Slide 49

Slide 49 text

(p < 10-3, d = 0.15) Goal-specific users browse less content, How does intent influence browsing? Goal-specific Goal-nonspecific Pins seen 156 144

Slide 50

Slide 50 text

(p < 10-3, effect size r = 0.35) and in fewer categories. Goal-specific users browse less content, How does intent influence browsing? Goal-specific Goal-nonspecific # distinct categories 10.4 8.7

Slide 51

Slide 51 text

(p < 10-3, χ2 = 20.8) Goal-specific users spend more time per session, and in fewer categories. Goal-specific users browse less content, How does intent affect browsing? Goal-specific Goal-nonspecific Sessions ≥ 30 mins 41.8% 48.0%

Slide 52

Slide 52 text

(p < 10-3, χ2 = 13.7) as do users with short-term goals! Goal-specific users spend more time per session, and in fewer categories. Goal-specific users browse less content, How does intent affect browsing? Short-term Long-term Sessions ≥ 30 mins 41.7% 48.4%

Slide 53

Slide 53 text

How does intent influence user retention? Does goal specificity increase the likelihood of return visits?

Slide 54

Slide 54 text

How does intent influence user retention? Do long-term goals increase the likelihood of return visits?

Slide 55

Slide 55 text

(p < 10-3, χ2 = 16.0 comparing day 0 and day 3) Goal-specific users are less likely to return soon, How does intent influence user retention? Returns 49% 53% 57% Days since initial visit Goal-nonspecific Goal-specific 0 1 2 3

Slide 56

Slide 56 text

(n.s.) but temporal range doesn’t affect future visits. Goal-specific users are less likely to return soon, How does intent influence user retention?

Slide 57

Slide 57 text

Intent × Behavior Goal Specificity Temporal Range Searching Browsing Saving Time spent Return visits

Slide 58

Slide 58 text

Intent × Behavior Goal Specificity Temporal Range Searching ˛* Browsing ▼* Saving Time spent ˛* Return visits ▼*

Slide 59

Slide 59 text

Intent × Behavior Goal Specificity Temporal Range Searching ˛* Browsing ▼* ˛* Saving ˛* Time spent ˛* ▼* Return visits ▼*

Slide 60

Slide 60 text

How does intent influence recipe-finding? Case study

Slide 61

Slide 61 text

(p < 0.01, d = 0.30) Users with short-term goals examine recipes closely, How does intent influence recipe-finding? Case study Short-term Long-term Recipe closeups 0.9 1.8

Slide 62

Slide 62 text

(p < 0.01, d = 0.28) and tend to be looking to make main courses. Users with short-term goals examine recipes closely, How does intent influence recipe-finding? Case study Short-term Long-term Recipes with meat or seafood 27% 42%

Slide 63

Slide 63 text

How does intent affect behavior? Can we predict intent (quickly)? 1 2 3 What is intent?

Slide 64

Slide 64 text

Can we predict intent of a user session?

Slide 65

Slide 65 text

Can we predict intent in the first ten minutes? Is the user acting in the long-term? Is the user goal-specific?

Slide 66

Slide 66 text

Can we predict intent in the first ten minutes? Demographics Current activity Historical activity gender, age, … searches, views, … past searches, past views, …

Slide 67

Slide 67 text

Can we predict intent in the first ten minutes? Demographics Historical Activity Current Activity 0.00 0.20 0.40 0.60 0.80 0.72 0.62 0.54 0.78 0.67 0.56 Goal Specificity Temporal Range AUC + +

Slide 68

Slide 68 text

Can we predict intent in the first ten minutes? Demographics Historical Activity Current Activity 0.00 0.20 0.40 0.60 0.80 0.72 0.62 0.54 0.78 0.67 0.56 Goal Specificity Temporal Range AUC + +

Slide 69

Slide 69 text

Can we predict intent in the first ten minutes? Demographics Historical Activity Current Activity 0.00 0.20 0.40 0.60 0.80 0.72 0.62 0.54 0.78 0.67 0.56 Goal Specificity Temporal Range AUC + +

Slide 70

Slide 70 text

Can we predict intent in the first ten minutes?

Slide 71

Slide 71 text

Can we predict intent in the first ten minutes? first minute?

Slide 72

Slide 72 text

Can we predict intent in the first ten minutes? 30 seconds? first minute?

Slide 73

Slide 73 text

Can we predict intent in the first ten minutes? 30 seconds (or less)? first minute? a majority of users have made 0 pins, closeups, searches

Slide 74

Slide 74 text

How quickly can we predict intent? AUC 0.6 0.65 0.7 0.75 0.8 Observation Period 0s 30s 1m 2m 3m 4m 5m 10m 15m 30m 45m 60m 180m Goal Specificity Temporal Range

Slide 75

Slide 75 text

How quickly can we predict intent? AUC 0.6 0.65 0.7 0.75 0.8 Observation Period 0s 30s 1m 2m 3m 4m 5m 10m 15m 30m 45m 60m 180m Goal Specificity Temporal Range

Slide 76

Slide 76 text

While 10 minutes is sufficient for predicting intent… AUC 0.6 0.65 0.7 0.75 0.8 Observation Period 0s 30s 1m 2m 3m 4m 5m 10m 15m 30m 45m 60m 180m Goal Specificity Temporal Range

Slide 77

Slide 77 text

…intent can be predicted much more quickly… AUC 0.6 0.65 0.7 0.75 0.8 Observation Period 0s 30s 1m 2m 3m 4m 5m 10m 15m 30m 45m 60m 180m Goal Specificity Temporal Range

Slide 78

Slide 78 text

…and can be predicted even before a session begins AUC 0.6 0.65 0.7 0.75 0.8 Observation Period 0s 30s 1m 2m 3m 4m 5m 10m 15m 30m 45m 60m 180m Goal Specificity Temporal Range

Slide 79

Slide 79 text

High-level intent can be predicted within minutes from low-level behavioral signals.

Slide 80

Slide 80 text

How do these findings apply?

Slide 81

Slide 81 text

Designing for browsing vs. doing How do these findings apply? task-based interfaces? prioritizing newness?

Slide 82

Slide 82 text

Situational recommendations and suggestions Designing for browsing vs. doing How do these findings apply? detecting changing intent specific or diverse?

Slide 83

Slide 83 text

Method adaptable to other content-based services Situational recommendations and suggestions Designing for browsing vs. doing How do these findings apply? Yelp/Zomato? Instagram?

Slide 84

Slide 84 text

Justin Cheng @jcccf, Caroline Lo @csuen, Jure Leskovec @jure / Stanford University + Pinterest http://bit.ly/pinterest-paper Predicting Intent Using Activity Logs How Goal Specificity and Temporal Range Affect User Behavior