Limits...
Gaze-Assisted User Intention Prediction for Initial Delay Reduction in Web Video Access.

Lee S, Yoo J, Han G - Sensors (Basel) (2015)

Bottom Line: The introduction of the sequential and concurrent flow of resources in human cognition and behavior can significantly improve the accuracy and preparation time for intention prediction.This paper introduces a threaded interaction model and applies it to user intention prediction for initial delay reduction in web video access.Experimental results show a 92% hit-ratio, 0.5-s initial delay on average and 1.5-s worst initial delay, which is much less than a user's tolerable limit in web video access, demonstrating significant improvement of accuracy and advance time in intention prediction by introducing the proposed threaded interaction model.

View Article: PubMed Central - PubMed

Affiliation: School of Integrated Technology, Yonsei University, Incheon 406-840, Korea. youb007@yonsei.ac.kr.

ABSTRACT
Despite the remarkable improvement of hardware and network technology, the inevitable delay from a user's command action to a system response is still one of the most crucial influence factors in user experiences (UXs). Especially for a web video service, an initial delay from click action to video start has significant influences on the quality of experience (QoE). The initial delay of a system can be minimized by preparing execution based on predicted user's intention prior to actual command action. The introduction of the sequential and concurrent flow of resources in human cognition and behavior can significantly improve the accuracy and preparation time for intention prediction. This paper introduces a threaded interaction model and applies it to user intention prediction for initial delay reduction in web video access. The proposed technique consists of a candidate selection module, a decision module and a preparation module that prefetches and preloads the web video data before a user's click action. The candidate selection module selects candidates in the web page using proximity calculation around a cursor. Meanwhile, the decision module computes the possibility of actual click action based on the cursor-gaze relationship. The preparation activates the prefetching for the selected candidates when the click possibility exceeds a certain limit in the decision module. Experimental results show a 92% hit-ratio, 0.5-s initial delay on average and 1.5-s worst initial delay, which is much less than a user's tolerable limit in web video access, demonstrating significant improvement of accuracy and advance time in intention prediction by introducing the proposed threaded interaction model.

No MeSH data available.


Initial delay and downlink bandwidth according to the number of preparation modules.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4507624&req=5

f10-sensors-15-14679: Initial delay and downlink bandwidth according to the number of preparation modules.

Mentions: The number of preparation modules showed a statistically significant (F(3,107) = 8.820 and p < 0.0005) influence on the initial delay, while the influence on the hit-ratio was insignificant (F(3,107) = 1.705, p = 0.170) by the ANOVA test. Figure 10 shows the initial delay, hit-ratio and downlink bandwidth per preparation module with respect to the number of preparation modules. The downlink bandwidth decreased as the number of preparation modules increased, showing an ā€˜Lā€™-shaped curve, because preparation modules shared limited total network bandwidth. Even though the initial delay decreased as the downlink bandwidth increased, the improvement was not significant when the downlink bandwidth exceeded a certain limit, because the server-client interaction time became a dominant portion of the initial delay. In the cursor-only case, despite the initial delay being improved as the number of preparation modules increased from one to two, an excess number of preparation modules degraded the initial delay. This means that an excessive number of preparation modules caused increased false decisions, which wasted network resources.


Gaze-Assisted User Intention Prediction for Initial Delay Reduction in Web Video Access.

Lee S, Yoo J, Han G - Sensors (Basel) (2015)

Initial delay and downlink bandwidth according to the number of preparation modules.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4507624&req=5

f10-sensors-15-14679: Initial delay and downlink bandwidth according to the number of preparation modules.
Mentions: The number of preparation modules showed a statistically significant (F(3,107) = 8.820 and p < 0.0005) influence on the initial delay, while the influence on the hit-ratio was insignificant (F(3,107) = 1.705, p = 0.170) by the ANOVA test. Figure 10 shows the initial delay, hit-ratio and downlink bandwidth per preparation module with respect to the number of preparation modules. The downlink bandwidth decreased as the number of preparation modules increased, showing an ā€˜Lā€™-shaped curve, because preparation modules shared limited total network bandwidth. Even though the initial delay decreased as the downlink bandwidth increased, the improvement was not significant when the downlink bandwidth exceeded a certain limit, because the server-client interaction time became a dominant portion of the initial delay. In the cursor-only case, despite the initial delay being improved as the number of preparation modules increased from one to two, an excess number of preparation modules degraded the initial delay. This means that an excessive number of preparation modules caused increased false decisions, which wasted network resources.

Bottom Line: The introduction of the sequential and concurrent flow of resources in human cognition and behavior can significantly improve the accuracy and preparation time for intention prediction.This paper introduces a threaded interaction model and applies it to user intention prediction for initial delay reduction in web video access.Experimental results show a 92% hit-ratio, 0.5-s initial delay on average and 1.5-s worst initial delay, which is much less than a user's tolerable limit in web video access, demonstrating significant improvement of accuracy and advance time in intention prediction by introducing the proposed threaded interaction model.

View Article: PubMed Central - PubMed

Affiliation: School of Integrated Technology, Yonsei University, Incheon 406-840, Korea. youb007@yonsei.ac.kr.

ABSTRACT
Despite the remarkable improvement of hardware and network technology, the inevitable delay from a user's command action to a system response is still one of the most crucial influence factors in user experiences (UXs). Especially for a web video service, an initial delay from click action to video start has significant influences on the quality of experience (QoE). The initial delay of a system can be minimized by preparing execution based on predicted user's intention prior to actual command action. The introduction of the sequential and concurrent flow of resources in human cognition and behavior can significantly improve the accuracy and preparation time for intention prediction. This paper introduces a threaded interaction model and applies it to user intention prediction for initial delay reduction in web video access. The proposed technique consists of a candidate selection module, a decision module and a preparation module that prefetches and preloads the web video data before a user's click action. The candidate selection module selects candidates in the web page using proximity calculation around a cursor. Meanwhile, the decision module computes the possibility of actual click action based on the cursor-gaze relationship. The preparation activates the prefetching for the selected candidates when the click possibility exceeds a certain limit in the decision module. Experimental results show a 92% hit-ratio, 0.5-s initial delay on average and 1.5-s worst initial delay, which is much less than a user's tolerable limit in web video access, demonstrating significant improvement of accuracy and advance time in intention prediction by introducing the proposed threaded interaction model.

No MeSH data available.