PREDICTING PREVALENCE OF INFLUENZALIKE ILLNESS FROM GEOTAGGED TWEETS

  • Slides: 24
Download presentation
PREDICTING PREVALENCE OF INFLUENZA-LIKE ILLNESS FROM GEO-TAGGED TWEETS ADVISER: JIA-LING, KOH SOURCE: WWW’ 17

PREDICTING PREVALENCE OF INFLUENZA-LIKE ILLNESS FROM GEO-TAGGED TWEETS ADVISER: JIA-LING, KOH SOURCE: WWW’ 17 SPEAKER: MING-CHIEH, CHIANG DATE: 2018/02/27 1

OUTLINE • Introduction • Method • Experiment • Conclusion 2

OUTLINE • Introduction • Method • Experiment • Conclusion 2

MOTIVATION 3

MOTIVATION 3

MOTIVATION • Modeling disease spread with Twitter data involves several challenges • Only relying

MOTIVATION • Modeling disease spread with Twitter data involves several challenges • Only relying on text classification can include large errors 4

GOAL • A strong positive linear correlation exists between the number of ILI-related tweets

GOAL • A strong positive linear correlation exists between the number of ILI-related tweets and the number of recorded influenza notifications at state scale 5

OUTLINE • Introduction • Method • Experiment • Conclusion 6

OUTLINE • Introduction • Method • Experiment • Conclusion 6

METHOD • Obtain a set of labeled training data • Apply a semi-supervised cascade

METHOD • Obtain a set of labeled training data • Apply a semi-supervised cascade learning approach to learning SVM classifiers • Distinguish “sick” tweets and “other” tweets 7

TRAINING SVM CLASSIFIERS 8

TRAINING SVM CLASSIFIERS 8

TRAINING SVM CLASSIFIERS • Two parameters: class weight and C parameter • Fix one

TRAINING SVM CLASSIFIERS • Two parameters: class weight and C parameter • Fix one parameter and vary the other within a wide range of values 9

CLASSIFICATION STAGE 10

CLASSIFICATION STAGE 10

CLASSIFICATION STAGE • Features: unigram, bigram, trigram • Example: “ I got the flu”

CLASSIFICATION STAGE • Features: unigram, bigram, trigram • Example: “ I got the flu” -> (i, get, flu, i get, get flu, i get flu) • Use TF-IDF features to represent tweet data 11

MODELING ILLNESS PREVALENCE • Fit a linear regression model • The number of influenza

MODELING ILLNESS PREVALENCE • Fit a linear regression model • The number of influenza notifications in each state: Internet access * Twitter penetration rate 12

OUTLINE • Introduction • Method • Experiment • Conclusion 13

OUTLINE • Introduction • Method • Experiment • Conclusion 13

EXPERIMENT • CSIRO • Geo-tagged tweets within Australia for the entire year of 2015

EXPERIMENT • CSIRO • Geo-tagged tweets within Australia for the entire year of 2015 • 8, 961, 932 tweets • 225, 641 unique users 14

EXPERIMENT • JSON format • Data cleaning and tokenization 15

EXPERIMENT • JSON format • Data cleaning and tokenization 15

EXPERIMENT • Classifier performance • 1027 tweets are found to be ILI-related 16

EXPERIMENT • Classifier performance • 1027 tweets are found to be ILI-related 16

EXPERIMENT • Temporal Analysis 17

EXPERIMENT • Temporal Analysis 17

EXPERIMENT • Spatial Analysis 18

EXPERIMENT • Spatial Analysis 18

EXPERIMENT • Regional Spatial Analysis 19

EXPERIMENT • Regional Spatial Analysis 19

EXPERIMENT 20

EXPERIMENT 20

EXPERIMENT 21

EXPERIMENT 21

LIMITATION • The scarcity of the tweets • The laboratory confirmed influenza notifications are

LIMITATION • The scarcity of the tweets • The laboratory confirmed influenza notifications are incomplete • Twitter is more popular among younger generations • Manual checking 22

OUTLINE • Introduction • Method • Experiment • Conclusion 23

OUTLINE • Introduction • Method • Experiment • Conclusion 23

CONCLUSION • Propose effective modifications to the state-ofart approach in detecting illness-related tweets •

CONCLUSION • Propose effective modifications to the state-ofart approach in detecting illness-related tweets • Twitter data is a reasonable proxy for detecting disease outbreak and possesses strong linear correlation with real-world influenza notification data 24

ILLNESS AND ILLNESS BEHAVIOUR INTRODUCTION Illness An illnessILLNESS AND ILLNESS BEHAVIOUR INTRODUCTION Illness An illness
U S Outpatient InfluenzaLike Illness Surveillance Network ILINetU S Outpatient InfluenzaLike Illness Surveillance Network ILINet
Community Validation of Influenzalike Illness as a PredictorCommunity Validation of Influenzalike Illness as a Predictor
Comparability of Electronic and Manual Influenzalike Illness ILIComparability of Electronic and Manual Influenzalike Illness ILI
Maintaining Canadas Flu Watch Influenzalike Illness ILI surveillanceMaintaining Canadas Flu Watch Influenzalike Illness ILI surveillance
Tweets Tweets for Asda Chosen By You sarahclarkeTweets Tweets for Asda Chosen By You sarahclarke
Mental Illness Mental Illness Also called emotional illnessMental Illness Mental Illness Also called emotional illness
Mental Illness What is mental illness Mental illnessMental Illness What is mental illness Mental illness
RECOMMENDING TRAVEL PLANS USING GEOTAGGED PHOTOS Harshitha RaoRECOMMENDING TRAVEL PLANS USING GEOTAGGED PHOTOS Harshitha Rao
The CAPi TA ILI study Monitoring of influenzalikeThe CAPi TA ILI study Monitoring of influenzalike
Forcing Seasonality of Influenzalike Epidemics with Daily SolarForcing Seasonality of Influenzalike Epidemics with Daily Solar
Smoking prevalence Smoking prevalence is high in TurkeySmoking prevalence Smoking prevalence is high in Turkey
Prevalence Age Sex and agespecific prevalence of migrainePrevalence Age Sex and agespecific prevalence of migraine
Prevalence of arterial hypertension worldwide Prevalence of arterialPrevalence of arterial hypertension worldwide Prevalence of arterial
EPIDEMIOLOGY General Prevalence of Joint Pain Prevalence ofEPIDEMIOLOGY General Prevalence of Joint Pain Prevalence of
EPIDEMIOLOGY General Prevalence of Acute Pain Lifetime prevalenceEPIDEMIOLOGY General Prevalence of Acute Pain Lifetime prevalence
Prevalence AFHF ATRIA study prevalence of AF inPrevalence AFHF ATRIA study prevalence of AF in
Ankylosing spondylitis Dr Chris Edwards Prevalence Worldwide prevalenceAnkylosing spondylitis Dr Chris Edwards Prevalence Worldwide prevalence
Predicting Human Papilloma Virus Prevalence and Vaccine PolicyPredicting Human Papilloma Virus Prevalence and Vaccine Policy
Integumentary System Wellness and Illness Wellness and IllnessIntegumentary System Wellness and Illness Wellness and Illness
Sudden Illness Sudden Illness Introduction Think back toSudden Illness Sudden Illness Introduction Think back to
Stress Illness CHAPTER 14 Stress Illness Stress theStress Illness CHAPTER 14 Stress Illness Stress the
Food Borne Illness Food Borne Illness n FoodFood Borne Illness Food Borne Illness n Food