Personalizing Threshold Values on Behavior Detection with Collaborative Filtering
Reference <ul><li>Hiroyuki Yamahara, Fumiko Harada, Hideyuki Takada,and Hiromitsu Shimakawa “ Personalizing Threshold Valu...
Outline <ul><li>Introduction </li></ul><ul><li>Behavior Detection in the Home </li></ul><ul><li>Discussion on Setting of T...
Introduction <ul><li>A collaboration system for assisting users in their homes, as an attempt for making intelligent envir...
Behavior Detection in the Home <ul><li>Detection of High-Level Bbehavior </li></ul><ul><li>Individual Habit in Touched Obj...
Detection of High-Level Behavior <ul><li>leaving the home, coming home, getting up and going to bed </li></ul><ul><li>cons...
Individual Habit in Touched Objects <ul><li>most people often have habitual actions in a habitual order, for not making om...
Behavior Detection with Ordered Pairs <ul><li>create a behavioral pattern represented by a set of  ordered pairs, which sh...
How to create a behavioral pattern
Discussion on Setting of Threshold Values <ul><li>Difficulty of Setting Threshold Values </li></ul><ul><li>Effect of Detec...
Difficulty of Setting Threshold Values <ul><li>TPR shows the rate at which behavior logs in a specific situation, which lo...
Effect of Detection Threshold on Recognition Accuracy of User Behavior <ul><li>Comparing the recognition rates in tables f...
Effect of Extraction Threshold on Recognition Accuracy of User Behavior
Dynamic Threshold Determination with Collaborative Filtering <ul><li>Dynamic Determination of Thresholds Suitable for Indi...
Dynamic threshold determination with collaborative filtering
Dynamic Determination of Thresholds Suitable for Individuals
Determination of Thresholds with Estimate Values by Collaborative Filtering
Related Work
Conclusion <ul><li>Realize the collaboration by providing proactive services according to user behavior </li></ul><ul><li>...
三種過濾方式之比較
Upcoming SlideShare
Loading in …5
×

Personalizing Threshold Values On Behavior Detection With Collaborative Filtering

725 views

Published on

Published in: Technology, Business
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
725
On SlideShare
0
From Embeds
0
Number of Embeds
20
Actions
Shares
0
Downloads
18
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Personalizing Threshold Values On Behavior Detection With Collaborative Filtering

  1. 1. Personalizing Threshold Values on Behavior Detection with Collaborative Filtering
  2. 2. Reference <ul><li>Hiroyuki Yamahara, Fumiko Harada, Hideyuki Takada,and Hiromitsu Shimakawa “ Personalizing Threshold Values on Behavior Detection with Collaborative Filtering ” Ubiquitous Intelligence and Computing 2008, LNCS 5061 </li></ul>
  3. 3. Outline <ul><li>Introduction </li></ul><ul><li>Behavior Detection in the Home </li></ul><ul><li>Discussion on Setting of Threshold Values </li></ul><ul><li>Dynamic Threshold Determination with Collaborative Filtering </li></ul><ul><li>Related Work </li></ul><ul><li>Conclusion </li></ul>
  4. 4. Introduction <ul><li>A collaboration system for assisting users in their homes, as an attempt for making intelligent environments </li></ul><ul><li>Collaboration with the user by environment may bring him comfort, relief and safety </li></ul><ul><li>proactive services </li></ul><ul><li>high-level behaviors </li></ul><ul><li>set threshold values, which are used for creating a behavioral pattern and for matching online sensor data with the pattern </li></ul>
  5. 5. Behavior Detection in the Home <ul><li>Detection of High-Level Bbehavior </li></ul><ul><li>Individual Habit in Touched Objects </li></ul><ul><li>Behavior Detection with Ordered Pairs </li></ul>
  6. 6. Detection of High-Level Behavior <ul><li>leaving the home, coming home, getting up and going to bed </li></ul><ul><li>consider that a high-level behavior is a long behavior of around ten minutes </li></ul><ul><li>characteristics of the highlevel behavior vary with individual user </li></ul>
  7. 7. Individual Habit in Touched Objects <ul><li>most people often have habitual actions in a habitual order, for not making omission of things to do, in situations such as leaving the home and going to bed </li></ul><ul><li>RFID tags </li></ul>
  8. 8.
  9. 9.
  10. 10. Behavior Detection with Ordered Pairs <ul><li>create a behavioral pattern represented by a set of ordered pairs, which show the order relation among touched objects, with histories of touched objects as sample behavior logs </li></ul>
  11. 11. How to create a behavioral pattern
  12. 12. Discussion on Setting of Threshold Values <ul><li>Difficulty of Setting Threshold Values </li></ul><ul><li>Effect of Detection Threshold on Recognition Accuracy of User Behavior </li></ul><ul><li>Effect of Extraction Threshold on Recognition Accuracy of User Behavior </li></ul>
  13. 13. Difficulty of Setting Threshold Values <ul><li>TPR shows the rate at which behavior logs in a specific situation, which logs are referred to as true cases, are correctly detected with a behavioral pattern of the situation </li></ul><ul><li>TNR shows the rate at which behavior logs in situations other than the specific situation, which logs are referred to as false cases, are correctly neglected with the behavioral pattern of the situation </li></ul>
  14. 14.
  15. 15.
  16. 16. Effect of Detection Threshold on Recognition Accuracy of User Behavior <ul><li>Comparing the recognition rates in tables from Table 1 to Table 4 with differences between common values and the best values of each subject in Table 5, it is apparent that the more differences bring lower recognition rate </li></ul>
  17. 17.
  18. 18. Effect of Extraction Threshold on Recognition Accuracy of User Behavior
  19. 19. Dynamic Threshold Determination with Collaborative Filtering <ul><li>Dynamic Determination of Thresholds Suitable for Individuals </li></ul><ul><li>Determination of Thresholds with Estimate Values by Collaborative Filtering </li></ul>
  20. 20. Dynamic threshold determination with collaborative filtering
  21. 21. Dynamic Determination of Thresholds Suitable for Individuals
  22. 22. Determination of Thresholds with Estimate Values by Collaborative Filtering
  23. 23. Related Work
  24. 24. Conclusion <ul><li>Realize the collaboration by providing proactive services according to user behavior </li></ul><ul><li>To detect user behavior precisely, our detection method dynamically determines threshold values suitable for behavioral patterns of individuals with collaborative filtering. </li></ul><ul><li>Our future work is evaluation of the determination method in experiments. </li></ul>
  25. 25.
  26. 26. 三種過濾方式之比較

×