臺灣博碩士論文加值系統

　　智慧型家庭是被認為是資訊產業的下一代。而情境感知系統正是實現無所不在、協助幫助人類提昇生活品質的關鍵方法。OSGi開放式家庭閘道框架是一個採用爪哇語言所創作跨平台的中介軟體，可以用來提供控制家用裝置的服務。由家庭成員與裝置間的互動了解其生活行為和所需的服務，並且讓家用閘道能夠自動發出控制訊號給相對應的裝置而達到自動化的效果是一件值得期待的事情。不過因為家庭成員個性的不同，對於同一件活動有著不同的方法或是順序達成，而且每個家庭所擁有的裝置數量和種類不一，專門為這些家庭訂做服務是很困難的。
　　因此本研究提出了一個能和OSGi框架複合在一起的框架，使其能夠維護安裝於OSGi上的知識，自動的偵測家庭的活動，並給予活動適當的協助，且能隨成員之習慣變化，進行自我維護、演變、回收過時的活動，讓學習模型針對每個家庭達到客制化的功效。本研究所提出的框架設計依照所提出的六相位學習流程來實現。完全相容於OSGi框架。
　　實驗結果也非常的樂觀，在一個會變異的動態系統中，複合框架在一星期之內即可累積出約百分之九十五左右的經驗給予下週使用，並且可以針對同一件活動的不確定性，以及每天生活的差異性進行演變、回收掉不需要的場景，以達節省系統資源、永續經營的目的。

　　Smart home is considered as one of next evolution of information technologies. Context aware ubiquitous computing provides the thought of invisible technology that helps, assists, and improves human’s quality of life. The OSGi framework is a cross platform service management middleware in home gateway. Services on OSGi framework controls home devices. Learning from human’s life patterns according to the interoperation of smart devices, discovering the scenarios and constructing the reasoning system are important issues in home automation. Due to the variety of the personalities, unpredictable variances of human decisions that achieve an activity, and unpredictable sequential device operations, to form and build an adaptive service is quite difficult and complex for service developers. For those reasons a self-services constructing learning framework is needed.
　　In this thesis, JSALearningFramework is proposed for knowledge maintenance, automatic scenarios recognition, learning procedures, and reasoning on the OSGi. The source and sink with state machine model ensures the safety of learning and reasoning procedures with guarded knowledge. The scenario garbage collector on JSALearningFramework also makes the model achieve the goal of evolution. Six phases of learning processes proposed makes JSALearningFramework complete through learning and reasoning.
　　The experiment result shows that with the garbage scenario collector enabled, JSALearningFramework keeps 10~20 potential composition services in smart home according to the operations collected from devices. The learning process approximates a stable number of scenarios that makes the system has the ability of stable reasoning. The model will be adaptive to newly formed behaviors by human according to the learning and garbage scenario collection process.
　　The major contributions of the thesis are the new JSALearningFramework, the co-operation framework with OSGi, and the garbage scenario collection mechanism for the least used scenarios. In addition, it can dynamically build a composite service, or services, for family members, according to their behaviors.

中文摘要 I
ABSTRACT II
ACKNOWLEDGEMENT III
CONTENT IV
LIST OF TABLES VI
LIST OF FIGURES VII
1. INTRODUCTION 1
2. GENERAL DISCUSSION ON LEARNING FRAMEWORK ON HOME GATEWAY 3
　2.1 OSGI PLATFORM 3
　2.2 APPLICATIONS ON CONTEXT AWARE UBIQUITOUS COMPUTING 4
　2.3 NEURAL NETWORK 6
　2.4 DATA MINING AND RULE DISCOVERY 7
　2.5 ONTOLOGY 7
3. GENERAL DISCUSSION ON LEARNING METHODOLOGIES AND CONTEXT AWARE SYSTEM 9
　3.1 PREVIOUS WORK FOR CONTEXT AWARE FRAMEWORK 9
　　3.1.1 Previous work on MIT Open Mind Common Sense, LifeNet, ConceptNet and EventNet 10
　　3.1.2 Methodologies on Learning Human Activities 11
　　3.1.3 Previous Works on OSGi Intelligent Home Gateway 12
　3.2 SCENARIOS ON INTELLIGENT HOUSE FRAMEWORK 13
4. PROPOSED INTELLIGENT LEARNING FRAMEWORK – JSALEARNINGFRAMEWORK 16
　4.1 PROPOSED INTELLIGENT HOUSE KEEPER REQUIREMENTS FOR OSGI 16
　4.2 DEALING WITH LOW LEVEL CONTEXT AND HIGH LEVEL CONTEXT ARCHITECTURE ON UPNP STAND-ALONE ARCHITECTURE AND OSGI BASED ARCHITECTURE 18
　4.3 ADVANCED ARCHITECTURE OF LEARNING MODEL: SOURCE AND SINK 20
　　Six Phases about Source and Sink Architecture 23
　　Phase 1: Device Interaction and Information Sliding Window 24
　　Phase 2: Sink Generation, or Merging 24
　　Phase 3: Device Control Life Cycle Learning 27
　　Phase 4: Painting the Meaning of the Sink 27
　　Phase 5: Garbage Collection 28
　　Phase 6: System Reasoning and Feedback Interaction 29
　4.4 DEALING SOURCE AND SINK WITH UPNP, INFORMATION RETRIEVING 35
　4.5 HUMAN TEMPORARY BEHAVIOR ELIMINATION 37
　4.6 SERVICE ARCHITECTURE AND DESIGN 39
　4.7 VOICE RECOGNITION AND USER FEEDBACK, HIGH LEVEL CONTEXT SOURCING 40
　4.8 DEVICE INTELLIGENCE AND AUTOMATIC ASSISTANT SAFETY ISSUE-SCENARIO GUARDING, PATTERN CORRECTNESS 41
　4.9 DEVELOPER’S CODING APIS AND SOFTWARE COMPONENTS 43
　4.10 SUMMERY 45
5. RESULT 47
6. CONCLUSION AND FUTURE WORKS 54
REFERENCE 56
APPENDIX 59
BIOGRAPHY 63

[1]　J. E. Bardram, "The Java Context Awareness Framework (JCAF) – A Service Infrastructure and Programming Framework for Context-Aware Applications," in Pervasive Computing, 2005, pp. 98-115.
[2]　O. Brdiczka, P. Reignier, J. L. Crowley, D. Vaufreydaz, and J. Maisonnasse, "Deterministic and probabilistic implementation of context," 2006, p. 5 pp.
[3]　L. Chia-Hsun Jackie, B. Leonardo, H. E. Jose, L. Henry, and S. Ted, "Augmenting kitchen appliances with a shared context using knowledge about daily events," in Proceedings of the 11th international conference on Intelligent user interfaces, Sydney, Australia, 2006, pp. 348-350.
[4]　J. L. Crowley, O. Brdiczka, and P. Reignier, "Learning Situation Models for Understanding Activity," in 5th International Conference on Development and Learning(ICDL06) Bloomington, Il., USA, 2006.
[5]　J. Espinosa and H. Lieberman, "EventNet: Inferring Temporal Relations Between Commonsense Events," in MICAI 2005: Advances in Artificial Intelligence, 2005, pp. 61-69.
[6]　T. Gu, H. K. Pung, and D. Q. Zhang, "Toward an OSGi-based infrastructure for context-aware applications," Pervasive Computing, IEEE, vol. 3, no. 4, pp. 66-74, 2004.
[7]　T. Gu, X. H. Wang, H. K. Pung, and D. Q. Zhang, "An Ontology-based Context Model in Intelligent Environments," in In Proceedings of Communication Networks and Distributed Systems Modeling and Simulation Conference (CNDS 2004), San Diego, California, USA, January, 2004, pp. 270-275.
[8]　H.-L. Hsu, "ContextSense–A Commonsense Knowledge Base System for Ambient Intelligence," in Architecture Tainan: National Cheng Kung University, 2006, p. 85.
[9]　JADE Framework - Java Agent DEvelopment Framework, available from http://jade.tilab.com/
[10]　Jena Framework, available from http://jena.sourceforge.net/
[11]　Jess, available from http://herzberg.ca.sandia.gov/jess/
[12]　C. Jonghwa, S. Dongkyoo, and S. Dongil, "Research and implementation of the context-aware middleware for controlling home appliances," Consumer Electronics, IEEE Transactions on, vol. 51, no. 1, pp. 301-306, 2005.
[13]　C. Jonghwa, S. Dongkyoo, and S. Dongil, "Research and implementation of the context-aware middleware for controlling home appliances," 2005, pp. 161-162.
[14]　R. Khosla, C. Lai, B. Chowdhury, and Q. Li, "Patterns of Behaviour Mediated by Cognitive Scripts and Emotional Attitudes - Context-Aware Engineering of Data Mining Systems," 2006, pp. 3-10.
[15]　K.-C. Lee, J.-H. Kim, J.-H. Lee, and K.-M. Lee, "Implementation of Ontology Based Context-Awareness Framework for Ubiquitous Environment," 2007, pp. 278-282.
[16]　H. Liu and P. Singh, "ConceptNet — A Practical Commonsense Reasoning Tool-Kit," BT Technology Journal, vol. 22, no. 4, pp. 211-226, 2004.
[17]　B. Morgan and P. Singh, "Elaborating sensor data using temporal and spatial commonsense reasoning," 2006, p. 4 pp.
[18]　M. C. Mozer, "The Neural Network House: An Environment that Adapts to its Inhabitants.," in Proceedings of the American Association for Artificial Intelligence Spring Symposium on Intelligent Environments. AAAI Press, 1998, pp. 110-114.
[19]　M. Muhlenbrock, O. Brdiczka, D. Snowdon, and J. L. Meunier, "Learning to detect user activity and availability from a variety of sensor data," 2004, pp. 13-22.
[20]　K. Nishigaki, K. Yasumoto, N. Shibata, M. Ito, and T. Higashino, "Framework and rule-based language for facilitating context-aware computing using information appliances," 2005, pp. 345-351.
[21]　B. Oliver, M. Jerome, and R. Patrick, "Automatic detection of interaction groups," in Proceedings of the 7th international conference on Multimodal interfaces Torento, Italy: ACM Press, 2005.
[22]　B. Oliver, R. Patrick, and L. C. James, "Automatic Development of an Abstract Context Model for an Intelligent Environment," in Proceedings of the Third IEEE International Conference on Pervasive Computing and Communications Workshops: IEEE Computer Society, 2005.
[23]　R. Oriana, F. Cristiano Di, R. Stefano, and R. Kimmo, "Unearthing Design Patterns to Support Context-Awareness," in Proceedings of the 4th annual IEEE international conference on Pervasive Computing and Communications Workshops, 2006, pp. 383-387.
[24]　OSGi Spec, available from http://www.osgi.org
[25]　C. A. Petri, "Kommunikation mit Automaten, PhD thesis," in Institut fur Instrumentelle Mathematik Bonn, 1962.
[26]　L. R. Rabiner, "A tutorial on hidden Markov models and selected applications in speech recognition," Proceedings of the IEEE, vol. 77, no. 2, pp. 257-286, 1989.
[27]　Cybergarage UPnP Library, available from http://www.cybergarage.org/
[28]　P. Singh, T. Lin, E. T. Mueller, G. Lim, T. Perkins, and W. L. Zhu, "Open Mind Common Sense: Knowledge Acquisition from the General Public," in On the Move to Meaningful Internet Systems 2002: CoopIS, DOA, and ODBASE : Confederated International Conferences CoopIS, DOA, and ODBASE 2002. Proceedings, 2002, pp. 1223-1237.
[29]　P. Singh and W. Williams, "LifeNet: A Propositional Model of Ordinary Human Activity.," in Proceedings of the Workshop on Distributed and Collaborative Knowledge Capture (DC-KCAP), Sanibel Island, FL, 2003.
[30]　M. Sven and R. Andry, "A survey of research on context-aware homes," in Proceedings of the Australasian information security workshop conference on ACSW frontiers 2003 - Volume 21, Adelaide, Australia, 2003, pp. 159-168.
[31]　S. Thomas, K. Kay, S. Frank, and Y. Ming, "Middleware Support for Context-Awareness in 4G Environments," in Proceedings of the 2006 International Symposium on on World of Wireless, Mobile and Multimedia Networks, 2006, pp. 203-211.
[32]　OWL, available from http://www.w3.org/TR/owl-features/
[33]　M. Weiser, "Building Invisible Interfaces - Slides for the keynote speech.," in User Interface, Systems, and Technologies (UIST) Conference, 1994.