現今由於雲端服務越來越普及，不僅僅是雲端服務供應商，其他產業的業者也開始蒐集使用者的資訊上傳至雲端進行儲存和分析，以期提供更好的使用者體驗服務，但蒐集大量資料的同時，要如何去保障使用者的隱私是一個相當重要的議題。
有研究提出以匿名方式進行使用者資訊保護，基於布隆過濾器來保障雲端環境下的使用者資訊，並且在匿名的同時還能夠提供查詢服務。但此機制有資料容納數量上的限制，且當資料負載量提高時，插入和搜尋資料的時間成本也會跟著提高，造成效能下降。
因此所以本研究提出Decay-Cuckoo Filter，在布穀鳥過濾器的雜湊表上做了一些改良，在原有的Bucket之後加上第二組Associate Bucket，用來儲存衰減係數(Decay Value)，一開始插入元素時，該元素的衰減係數為最大值，而此係數會隨著時間的經過、插入元素數量或者查詢元素素量達設定值後執行「衰減」，當某元素衰減係數衰減至最小值“0”，該元素將從資料表中被刪除，釋放空間。且本研究實驗證明，使用Decay Function方法其元素插入時間效能提升了1.4倍，元素的查詢時間效能提升了2倍。

Nowadays, due to the increasing popularity of cloud services, not only cloud service providers, but also industry players in other industries have begun to collect user information and upload it to the cloud for storage and analysis, in order to provide better user experience services, but collect a large amount of data. At the same time, how to protect the privacy of users is a very important issue.
Some studies have proposed to protect user information in an anonymous manner. Based on the Bloom filter, user information in the cloud environment is guaranteed, and the query service can be provided while being anonymous. However, this mechanism has a limitation on the number of data storage, and as the data load increases, the time cost of inserting and searching for data will also increase, resulting in a decrease in performance.
Therefore, this study proposes a Decay Value as the usage weight value of each data, and a set of Decay Function to let the attenuation value decay with the custom conditions, when the attenuation value is reduced to At the minimum value, the data is removed from the data sheet to release the location, extending the life of the data sheet and improving its performance. And this research experiment proves that using the Decay Function method, the element insertion time is increased by 1.4 times, and the element query time is increased by 2 times.

[參考網站]
[1] Bloom Filters the math Retrieved (Accessed：20-Jun-2018) From http://pages.cs.wisc.edu/~cao/papers/summary-cache/node8.html (Accessed：20-Jun-2017)
[2] EUGDPR .org “Home Page of EU GDPR” (Accessed：20-Jun-2018) From https://www.eugdpr.org
[3] The Next Web(2017). (Accessed：20-Jun-2018) From “Android malware spreads like wildfire: 350 new malicious APPs every hour” from https://tnw.to/2pBScst.
[4] TrendMicro(2016). (Accessed：20-Jun-2018) From “Information on Reported Vulnerabilities in Trend Micro Password Manager” form https://blog.trendmicro.com/information-on-reported-vulnerabilities-in-trend-micro-password-manager/
[5] We are social(2017). (Accessed：20-Jun-2018) From "DIGITAL IN 2017: GLOBAL OVERVIEW"from https://wearesocial.com/special-reports/digital-in-2017-global-overview.
[6] 經濟部標準檢驗局.“個人資料去識別化適用標準” (Accessed：20-Jun-2018) Fromhttps://www.bsmi.gov.tw/wSite/public/Data/f1447142556627.pdf
[中文文獻]
[7] 張瑋倫.“基於Bloom Filter之雲端環境下使用者匿名機制查詢研究.”中央大學資訊管理學系所碩士論文 (2017)
[8] 曹宗恆.“混合式硬碟緩衝快取區管理機制.” 交通大學資訊管理學系所碩士論文 (2006)
[英文文獻]
[9] Abdelhameed, S. A., Moussa, S. M., & Khalifa, M. E. (2017, December). Enhanced additive noise approach for privacy-preserving tabular data publishing. In Intelligent Computing and Information Systems (ICICIS), 2017 Eighth International Conference on (pp. 284-291). IEEE.
[10] Bloom, B. H. (1970). Space/time trade-offs in hash coding with allowable errors. Communications of the ACM, 13(7), 422-426.
[11] Bao, S., Hathal, W., Cruickshank, H., Sun, Z., Asuquo, P., & Lei, A. (2017). A lightweight authentication and privacy-preserving scheme for VANETs using TESLA and Bloom Filters. ICT Express.
[12] Charu, A., & Philip, S. Y. (2008). Privacy-preserving data mining: Models and algorithms.
[13] Coffman, E. G., & Denning, P. J. (1973). Operating systems theory (Vol. 973). Englewood Cliffs, NJ: Prentice-Hall.
[14] Damopoulos, D., Kambourakis, G., & Portokalidis, G. (2014, April). The best of both worlds: a framework for the synergistic operation of host and cloud anomaly-based IDS for smartphones. In Proceedings of the Seventh European Workshop on System Security (p. 6). ACM.
[15] Domingo-Ferrer, J., & Torra, V. (2005). Ordinal, continuous and heterogeneous k-anonymity through microaggregation. Data Mining and Knowledge Discovery, 11(2), 195-212.
[16] Fan, B., Andersen, D. G., Kaminsky, M., & Mitzenmacher, M. D. (2014, December). Cuckoo filter: Practically better than bloom. In Proceedings of the 10th ACM International on Conference on emerging Networking Experiments and Technologies (pp. 75-88). ACM.
[17] Fan, L., Cao, P., Almeida, J., & Broder, A. Z. (2000). Summary cache: a scalable wide-area web cache sharing protocol. IEEE/ACM transactions on networking, 8(3), 281-293.
[18] Fountoulakis, N., Khosla, M., & Panagiotou, K. (2016). The multiple-orientability thresholds for random hypergraphs. Combinatorics, Probability and Computing, 25(6), 870-908.
[19] Kuroiwa, K., & Uda, R. (2012, February). A low cost privacy protection method for SNS by using bloom filter. In Proceedings of the 6th International Conference on Ubiquitous Information Management and Communication (p. 7). ACM.
[20] Machanavajjhala, A., Gehrke, J., Kifer, D., & Venkitasubramaniam, M. (2006, April). \ell-Diversity: Privacy Beyond\kappa-Anonymity. In null (p. 24). IEEE.
[21] Pagh, R., & Rodler, F. F. (2004). Cuckoo hashing. Journal of Algorithms, 51(2), 122-144.
[22] Pan, Y., Zhu, X. L., & Chen, T. G. (2012). Research on Privacy Preserving on K-anonymity. Journal of Software, 7(7), 1649-1656.
[23] Rashidi, B., Fung, C., & Vu, T. (2014, September). Recdroid: A resource access permission control portal and recommendation service for smartphone users. In Proceedings of the ACM MobiCom workshop on Security and privacy in mobile environments (pp. 13-18). ACM.
[24] Rottenstreich, O., Kanizo, Y., & Keslassy, I. (2014). The variable-increment counting Bloom filter. IEEE/ACM Transactions on Networking (TON), 22(4), 1092-1105.
[25] Shen, N., Yang, J., Yuan, K., Fu, C., & Jia, C. (2016). An efficient and privacy-preserving location sharing mechanism. Computer Standards & Interfaces, 44, 102-109.
[26] Smith, D. (2017). Secure pseudonymisation for privacy-preserving probabilistic record linkage. Journal of Information Security and Applications, 34, 271-279.
[27] Truong, H. T. T., Lagerspetz, E., Nurmi, P., Oliner, A. J., Tarkoma, S., Asokan, N., & Bhattacharya, S. (2014, April). The company you keep: Mobile malware infection rates and inexpensive risk indicators. In Proceedings of the 23rd international conference on World wide web (pp. 39-50). ACM.
[28] Vatsalan, D., & Christen, P. (2016). Multi-Party Privacy-Preserving Record Linkage using Bloom Filters. arXiv preprint arXiv:1612.08835.
[29] Vatsalan, D., & Christen, P. (2016). Privacy-preserving matching of similar patients. Journal of biomedical informatics, 59, 285-298.
[30] Wang, C. Y., Yang, W. P., Tseng, J. C., & Hsu, M. (1990). Random filter and its analysis. International journal of computer mathematics, 33(3-4), 181-194.
[31] Xie, K., Wang, X., Li, W., Zheng, Z., Xie, G., & Wen, J. (2016, June). Bloom-filter-based profile matching for proximity-based mobile social networking. In Sensing, Communication, and Networking (SECON), 2016 13th Annual IEEE International Conference on (pp. 1-9). IEEE.