RESUMO
Collaborative filtering has been widely used in recommendation systems to recommend items that users might like. However, collaborative filtering based recommendation systems are vulnerable to shilling attacks. Malicious users tend to increase or decrease the recommended frequency of target items by injecting fake profiles. In this paper, we propose a Kalman filter-based attack detection model, which statistically analyzes the difference between the actual rating and the predicted rating calculated by this model to find the potential abnormal time period. The Kalman Filter filters out suspicious ratings based on the abnormal time period and identifies suspicious users based on the source of these ratings. The experimental results show that our method performs much better detection performance for the shilling attack than the traditional methods.
RESUMO
OBJECTIVE@#To study the value of flow cytometric scoring system in the diagnosis of myelodysplastic syndromes (MDS).@*METHODS@#The phenotypes of erythroid and immature cells were analyzed retrospectively in 130 MDS patients, 19 healthy controls and 89 pathological controls, all of them were well clinically immunophenotyped. The 4-parameter scoring system reported in the literature was studied, including myeloblast-related cluster size, B-progenitor-related cluster size, lymphocyte to myeloblast CD45 ratio, and granulocyte to lymphocyte side scatter ratio. The two flow cytomatric parameters of the erythroid scoring system were analyzed, including CD36 coefficient of variation (CV) and CD71CV. According to our previous study, the percentage of CD117CD105 myeloid progenitor cells and the proportion of CD105 cells in CD117 cells were selected to establish a two-parameter scoring system, and compared with the four-parameter scoring system and the erythroid scoring system.@*RESULTS@#The sensitivity of the four-parameter scoring system and the erythroid scoring system for the diagnosis of low-risk MDS was 43.5% and 63.0%, and the specificity was 87.0% and 63.9%, respectively. After combining the two scoring systems, the sensitivity to diagnose low-risk MDS was 73.9% and the specificity was 62.0%. The sensitivity of the two-parameter scoring system for the diagnosis of low-risk MDS was 76.1% with a specificity of 81.5%. Combined with the four-parameter scoring system, the sensitivity was increased to 78.3%, but the specificity was reduced to 71.3%. After combining with the erythroid scoring system, the sensitivity reached 87.0%, but the specificity was reduced to 54.6%.@*CONCLUSION@#Using the two-parameter scoring system alone can achieve great sensitivity and specificity in the diagnosis of low risk MDS.