ESG Finance Week 2012

Analysis and Forecasting of Trending Topics in Online Media

Master's Thesis Defense

Tim AlthoffGerman Research Center for Artificial Intelligence (DFKI)

Analysis and Forecasting of Trending Topics in Online Media 2

Motivation

Analysis and Forecasting of Trending Topics in Online Media 3

Large-scale Record of Human Behavior

• Primary method for communication by college students in the U.S.

• 340 million tweets each day by 500 million users

• One billion search requests and about twenty petabytes of user-generated data each day (2009)

Analysis and Forecasting of Trending Topics in Online Media 4

Motivation

• People use web for news, information, and communication

• Online activity becomes indicative of the interests of the global population

• After Boston, e.g. Twitter has become a trusted channelfor authorities

› „Mirror of Society“[Steven Van Belleghem, 2012]

Analysis and Forecasting of Trending Topics in Online Media 5

Use Case of Online Activity

Analysis and Forecasting of Trending Topics in Online Media 6

Tracking Flu Infections

flu remedy

[Ginsberg et al. Detecting influenza epidemics using search engine query data. Nature. 2008.]

Analysis and Forecasting of Trending Topics in Online Media 7

Shortcomings of Related Work

1. Focus on very specific use cases (e.g. flu)2. Nowcasting instead of forecasting3. Manual correlation analysis instead of fully

automatic forecasting4. No comparison/study of different online & social

media channels wrt. emerging trends

[Choi and Varian. Predicting initial claims for unemployment benefits. Google, Inc. 2009.][Choi and Varian. Predicting the Present with Google Trends. Economic Record, 2012.][Ginsberg et al. Detecting influenza epidemics using search engine query data. Nature. 2008.][Cooper et al. Cancer internet search activity on a major search engine. Journal of medical Internet research, 2005.][Goel et al. Predicting consumer behavior with web search. National Academy of Sciences, 2010.][Jin et al. The wisdom of social multimedia: using Flickr for prediction and forecast. ACM MM 2010.][Adar et al. Why we search: visualizing and predicting user behavior. WWW, 2007.]

Analysis and Forecasting of Trending Topics in Online Media 8

Goals of Master's Thesis

1. Multi-channel analysis of trending topics

2. Fully automatic forecasting of trending topics

Analysis and Forecasting of Trending Topics in Online Media 9

Trending Topics

Analysis and Forecasting of Trending Topics in Online Media 10

Trending TopicsA trending topic• is a subject matter of discussion agreed on by a group of

people• experiences a sudden spike in user interest or engagement

Analysis and Forecasting of Trending Topics in Online Media 11

• Daily crawling of 10 sources• Capturing

• Communication needs• Search pattern• Information demand

Trending Topics Discovery

Trends

Search Germany

Search USA

News Germany

News USA

English Wikipedia

German Wikipedia

Trends Germany

Trends USA

Daily Trends

[Damian Borth, Adrian Ulges, Thomas Breuel. Dynamic Vocabularies for Web-based Concept Detection by Trend Discovery. ACM Multimedia, 2012]

Analysis and Forecasting of Trending Topics in Online Media 12

• Daily crawling of 10 sources• Capturing

• Communication needs• Search pattern• Information demand

Trending Topics Discovery

Trends

Search Germany

Search USA

News Germany

News USA

English Wikipedia

German Wikipedia

Trends Germany

Trends USA

Daily Trends

Dataset• Observation period– Sept. 2011 – Sept. 2012

• 40,000 items analyzed• ~3000 trending topics discovered

[Damian Borth, Adrian Ulges, Thomas Breuel. Dynamic Vocabularies for Web-based Concept Detection by Trend Discovery. ACM Multimedia, 2012]

Analysis and Forecasting of Trending Topics in Online Media 13

Trending Topics Discovery

Analysis and Forecasting of Trending Topics in Online Media 14

Trending Topics Discovery

Analysis and Forecasting of Trending Topics in Online Media 15

Multi-Channel Analysis

Analysis and Forecasting of Trending Topics in Online Media 16

Example

Neil Armstrong died on August 25, 2012.

Analysis and Forecasting of Trending Topics in Online Media 17

Lifetime Analysis

Analysis and Forecasting of Trending Topics in Online Media 18

Topic Category Analysis

Analysis and Forecasting of Trending Topics in Online Media 19

Topic Category Analysis

Analysis and Forecasting of Trending Topics in Online Media 20

Topic Category Analysis

Analysis and Forecasting of Trending Topics in Online Media 21

Topic Category Analysis

Analysis and Forecasting of Trending Topics in Online Media 22

Forecasting

Analysis and Forecasting of Trending Topics in Online Media 23

Forecasting

time

observed values forecast

Analysis and Forecasting of Trending Topics in Online Media 24

Basic Forecasting TechniquesNaive

Linear Trend

Average/Median

ARMA

Nearest Neighbor

Analysis and Forecasting of Trending Topics in Online Media 25

Basic Forecasting TechniquesNaive

Linear Trend

Average/Median

ARMA

Nearest Neighbor

Analysis and Forecasting of Trending Topics in Online Media 26

Basic Forecasting TechniquesNaive

Linear Trend

Average/Median

ARMA

Nearest Neighbor

Analysis and Forecasting of Trending Topics in Online Media 27

Basic Forecasting TechniquesNaive

Linear Trend

Average/Median

ARMA

Nearest Neighbor

Analysis and Forecasting of Trending Topics in Online Media 28

Basic Forecasting TechniquesNaive

Linear Trend

Average/Median

ARMA

Nearest Neighbor

Analysis and Forecasting of Trending Topics in Online Media 29

Example Signal

Analysis and Forecasting of Trending Topics in Online Media 30

Forecasting Can Be Very Challenging

Analysis and Forecasting of Trending Topics in Online Media 31

Recurring Signal

Analysis and Forecasting of Trending Topics in Online Media 32

Recurring Signal

Analysis and Forecasting of Trending Topics in Online Media 33

What To Do In This Case?

Analysis and Forecasting of Trending Topics in Online Media 34

What To Do In This Case?

Semantics!

Analysis and Forecasting of Trending Topics in Online Media 35

Exploiting Semantically Similar Topics

Analysis and Forecasting of Trending Topics in Online Media 36

Approach

Analysis and Forecasting of Trending Topics in Online Media 37

Approach

Analysis and Forecasting of Trending Topics in Online Media 38

Approach

Analysis and Forecasting of Trending Topics in Online Media 39

Discovering Semantically Similar Topics

Trending Topic:“Olympics 2012”

Analysis and Forecasting of Trending Topics in Online Media 40

Nearest Neighbor Sequence Matching

Analysis and Forecasting of Trending Topics in Online Media 41

Forecasting

Analysis and Forecasting of Trending Topics in Online Media 42

Forecasting

Analysis and Forecasting of Trending Topics in Online Media 43

Forecasting

Analysis and Forecasting of Trending Topics in Online Media 44

Forecasting

Analysis and Forecasting of Trending Topics in Online Media 45

Forecasting

Analysis and Forecasting of Trending Topics in Online Media 46

Forecasting

Analysis and Forecasting of Trending Topics in Online Media 47

Evaluation

Analysis and Forecasting of Trending Topics in Online Media 48

Dataset

• Public dataset of Wikipedia view statistics• Historical data from 2008 until today• Large-scale: 5 million articles attracting

870 million views each day – 2.8 TB compressed

→ Semantic similarity between 5 million articles→ 9 billion sequences to choose from

• Use 200 top trending topics for evaluationThanks Jörn!

Analysis and Forecasting of Trending Topics in Online Media 49

Discovering Semantically Similar Topics

Analysis and Forecasting of Trending Topics in Online Media 50

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 51

Example: “The Hunger Games”

Trending TopicTrigger

14 day horizonTime in days

Wikipedia article views

Different models

Analysis and Forecasting of Trending Topics in Online Media 52

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 53

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 54

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 55

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 56

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 57

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 58

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 59

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 60

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 61

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 62

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 63

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 64

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 65

Example: “The Hunger Games”

Analysis and Forecasting of Trending Topics in Online Media 66

Quantitative Results

Analysis and Forecasting of Trending Topics in Online Media 67

Root Mean Square Error

Analysis and Forecasting of Trending Topics in Online Media 68

Root Mean Square Error

Analysis and Forecasting of Trending Topics in Online Media 69

Root Mean Square Error

Analysis and Forecasting of Trending Topics in Online Media 70

Mean Average Percentage Error

Analysis and Forecasting of Trending Topics in Online Media 71

Mean Average Percentage Error

Analysis and Forecasting of Trending Topics in Online Media 72

Mean Average Percentage Error

Analysis and Forecasting of Trending Topics in Online Media 73

Conclusion

Analysis and Forecasting of Trending Topics in Online Media 74

Main Contributions

1. Multi-channel analysis of trending topics2. Fully automatic forecasting technique exploiting

semantic relationships between topics3. Empirical evaluation on a large-scale Wikipedia

dataset

Analysis and Forecasting of Trending Topics in Online Media 75

Insights

1. Trends on Twitter and Wikipedia are significantly more ephemeral than on Google.

2. All observed media channels tend to specialize in specific topic categories.

3. Semantically similar topics exhibit similar behavior.4. Exploiting this observation can significantly improve

forecasting performance (by 20-90% compared to baselines).

Analysis and Forecasting of Trending Topics in Online Media 76

Acknowledgments

Tim Althoff, Damian Borth, Jörn Hees, and Andreas Dengel.Analysis and Forecasting of Trending Topics in Online Media Streams. Submitted to ACM Multimedia, 2013.

Analysis and Forecasting of Trending Topics in Online Media 77

Questions?

Thank you for your attention!

Analysis and Forecasting of Trending Topics in Online Media 78

Backup Slides

Analysis and Forecasting of Trending Topics in Online Media 79

Motivation

• Big Data: “High volume, high velocity, and/or high variety information assets that require new forms of processing to enable enhanced decision making, insight discovery and process optimization”

• Large-scale user behavior

› „Mirror of Society“

[Mark A. Beyer and Douglas Laney. The importance of ’big data’: A definition. June 2012.]

Analysis and Forecasting of Trending Topics in Online Media 80

Large-scale Record of Human Behavior

• 1.2 million video minutes per second (est. 2016)

• Primary method for communication by college students in the U.S.

• 25 million articles and 1.8 billion page edits in 285 languages created by 100k contributors

• 340 million tweets each day by 500 million users

• One billion search requests and about twenty petabytes of user-generated data each day (2009)

Analysis and Forecasting of Trending Topics in Online Media 81

Unemployment Claims

jobs

[Choi and Varian. Predicting initial claims for unemployment benefits. Google, Inc. 2009.]

Analysis and Forecasting of Trending Topics in Online Media 82

Political Elections

US Democratic Primaries 2008hillary

[Jin et al. The wisdom of social multimedia: using Flickr for prediction and forecast. ACM Multimedia 2010.]

Analysis and Forecasting of Trending Topics in Online Media 83

What is the world talking about?Social Media can offer insights in• what the world is talking about• how people feel about events, brands and productsresulting in an collective awareness for trending topics

A trending topic• is a subject matter of discussion

agreed on by a group of people• experiences a sudden spike in

user interest or engagement

[Damian Borth, Adrian Ulges, Thomas Breuel. Dynamic Vocabularies for Web-based Concept Detection by Trend Discovery. ACM Multimedia, 2012]

Analysis and Forecasting of Trending Topics in Online Media 84

Trending Topics Discovery

Analysis and Forecasting of Trending Topics in Online Media 85

Trending Topics Discovery

Analysis and Forecasting of Trending Topics in Online Media 86

Lifetime Analysis

Analysis and Forecasting of Trending Topics in Online Media 87

Lifetime Analysis

Analysis and Forecasting of Trending Topics in Online Media 88

Lifetime Analysis

Analysis and Forecasting of Trending Topics in Online Media 89

Lifetime Analysis

Analysis and Forecasting of Trending Topics in Online Media 90

Basic Forecasting TechniquesNaive

Linear Trend

Average/Median

ARMA

Nearest Neighbor

Analysis and Forecasting of Trending Topics in Online Media 91

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 92

Example 2: “Battlefield 3”

Trending TopicTrigger

14 day horizonTime in days

Wikipedia article views

Analysis and Forecasting of Trending Topics in Online Media 93

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 94

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 95

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 96

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 97

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 98

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 99

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 100

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 101

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 102

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 103

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 104

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 105

Example 2: “Battlefield 3”

Analysis and Forecasting of Trending Topics in Online Media 106

Example 2: “Battlefield 3”