Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu - - PowerPoint PPT Presentation

▶

Feb 13, 2024 462 likes •675 views

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 2017-11-27 Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 1/19 Introduction Motivations Interruptions divided into self-interruptions and external

SLIDE 1

Detecting Self-Interruptions during Reading

Jan Pilzer and Sam Liu 2017-11-27

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 1/19

SLIDE 2

Introduction

Motivations

◮ Interruptions divided into self-interruptions and external

interruptions

◮ Self-interruptions are more costly ◮ Self-interruptions can originate from a loss of focus ◮ Self-interruptions should be detectable using biometric sensors

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 2/19

SLIDE 3

Introduction

Research Questions

◮ (RQ1) Does the low-cost eye tracker have the sufficient

accuracy to track the current reading line?

◮ (RQ2) Does eye gaze behave differently before a

self-interruption?

◮ (RQ3) Can we detect eye gaze patterns that occur before a

self-interruption?

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 3/19

SLIDE 4

Experiment

Setup

Figure: GazeReader setup using the Tobii Eye Tracker 4C after the participant didn’t look at the screen for some time.

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 4/19

SLIDE 5

Analysis

Session data

Figure: Example session data around a distraction event.

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 5/19

SLIDE 6

Analysis

Data segmentation

⎧ ⎪ ⎨ ⎪ ⎩ ti

⎧ ⎪ ⎨ ⎪ ⎩

tr1 BLUR FOCUS

Stripped data During normal reading

Before self-interruption

(non-reading related activities)

tr2 BLUR FOCUS

(reading related activities)

⎧ ⎨ ⎩

tp

⎧ ⎨ ⎩

Figure: Data segmentation scenario: Switch to another window which is reading related (top) or self interruption (bottom).

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 6/19

SLIDE 7

Analysis

Data segmentation (continued)

⎧ ⎪ ⎨ ⎪ ⎩ ⎧ ⎨ ⎩ ti tw

⎧ ⎪ ⎨ ⎪ ⎩

tr1

Look away Look away Look back

Pop-up 1

Stripped data During normal reading

Before self-interruption

(select “self-interruption”)

⎧ ⎨ ⎩ tw

⎧ ⎪ ⎪ ⎪ ⎨ ⎩

tr1 tp

Look back

Pop-up (select “external interruption/take note”)

⎧ ⎨ ⎩

external interruption reading related

tr2 ⎧ ⎨ ⎩

Figure: Data segmentation scenario: Looking away from the screen due to self interruption (top), or external interruption or reading related tasks (bottom).

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 7/19

SLIDE 8

Analysis

Features

Table: Features

Features Sub features fixation duration mean median variance min max count saccade duration mean median variance min max length mean median variance min max angle mean median variance min max

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 8/19

SLIDE 9

Analysis

Classification

(Average over 5 runs) Data for (ti) Training Data Testing Data SVM K nearest neighbors Logistic regression …

1/3 2/3

Training Features Training Labels Testing Features Testing Labels Predicted Labels Calculate metrics (SVM) Calculate metrics (Knn) Calculate metrics (LR) Calculate metrics (…)

Figure: Process of classifications for a (ti)

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 9/19

SLIDE 10

Analysis

Metrics

Table: Confusion matrix. positive: self-interruption, negative: concentrating on reading

Predicted positive Predicted negative Labeled positive True positive :-) False negative :-| Labeled negative False positive :-( True negative :-|

◮ Accuracy = true all ◮ Precision = true positive predicted positive ◮ Recall = true positive labeled positive ◮ F1 score = 2 · precision·recall precision+recall

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 10/19

SLIDE 11

Analysis

Classifiers

1. AdaBoostClassifier
2. DecisionTreeClassifier
3. GaussianNB
4. GaussianProcessClassifier
5. KNeighborsClassifier
6. LogisticRegression
7. MLPClassifier
8. QuadraticDiscriminantAnalysis
9. RandomForestClassifier
10. SVC (non-linear)
11. SVC (linear)

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 11/19

SLIDE 12

Results

Accuracy

10 20 30 40 50 60 ti/s 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 accuracy

AdaBoost DecisionTree GaussianNB KNeighbors LogisticRegression MLP QuadraticDiscriminantAnalysis RandomForest SVC (linear)

Figure: Accuracy scores of different tis and classifiers. Each line represents the result of a classifier.

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 12/19

SLIDE 13

Results

Precision

10 20 30 40 50 60 ti/s 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 precision

AdaBoost DecisionTree GaussianNB KNeighbors LogisticRegression MLP QuadraticDiscriminantAnalysis RandomForest SVC (linear)

Figure: Precision scores of different tis and classifiers. Each line represents the result of a classifier.

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 13/19

SLIDE 14

Results

Recall

10 20 30 40 50 60 ti/s 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 recall

AdaBoost DecisionTree GaussianNB KNeighbors LogisticRegression MLP QuadraticDiscriminantAnalysis RandomForest SVC (linear)

Figure: Recall scores of different tis and classifiers. Each line represents the result of a classifier.

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 14/19

SLIDE 15

Results

F1 Score

10 20 30 40 50 60 ti/s 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 f1

AdaBoost DecisionTree GaussianNB KNeighbors LogisticRegression MLP QuadraticDiscriminantAnalysis RandomForest SVC (linear)

Figure: F1 scores of different tis and classifiers. Each line represents the result of a classifier.

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 15/19

SLIDE 16

Conclusion

Contributions

◮ We prove that before a self-interruption, the eye movement is

different compared to when the reader concentrates on reading.

◮ We demonstrate with proper classifiers, we can detect the

incoming self-interruptions by gaining 5-second eye movement data before the interruption event.

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 16/19

SLIDE 17

Conclusion

Limitations

◮ Insufficient data, some participants rarely self-interrupt. ◮ Many participants usually prefer to read the print paper. ◮ Tobii 4C driver only runs on Windows and not accurate

enough.

◮ Classification is binary. ◮ Do not know which features weight more.

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 17/19

SLIDE 18

Conclusion

Future work

◮ How long in advance can we predict? ◮ Use pupil dilation metrics as features. ◮ Analyzing sentences/words: functional words vs content

words.

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 18/19

SLIDE 19

Q&A

Questions?

Detecting Self-Interruptions during Reading Jan Pilzer and Sam Liu 19/19