A new recommendation method was presented based on memetic algorithm-based clustering. The proposed method was tested on four highly sparse real-world datasets. Its recommendation performance is evaluated and compared...A new recommendation method was presented based on memetic algorithm-based clustering. The proposed method was tested on four highly sparse real-world datasets. Its recommendation performance is evaluated and compared with that of the frequency-based, user-based, item-based, k-means clustering-based, and genetic algorithm-based methods in terms of precision, recall, and F1 score. The results show that the proposed method yields better performance under the new user cold-start problem when each of new active users selects only one or two items into the basket. The average F1 scores on all four datasets are improved by 225.0%, 61.6%, 54.6%, 49.3%, 28.8%, and 6.3% over the frequency-based, user-based, item-based, k-means clustering-based, and two genetic algorithm-based methods, respectively.展开更多
From the view of both objective and subjective factors,the indoor air quality(IAQ)evaluation was considered.Carbon dioxide(CO2)and formaldehyde(HCHO)were selected as the typical contaminants of indoor air,and the eval...From the view of both objective and subjective factors,the indoor air quality(IAQ)evaluation was considered.Carbon dioxide(CO2)and formaldehyde(HCHO)were selected as the typical contaminants of indoor air,and the evaluation method of logarithmic index was adopted as the evaluation means of IAQ.Then the recommended limits(RL)of typical contaminants CO2 and HCHO were given through analysis and calculation.The limits of CO2 and HCHO in Indoor Air Quality Standard of China or other existing standards probably correspond to the level of PD=25(%).The result shows that the existing standards fail to meet the requirement of the definition of "acceptable indoor air quality",that is to say,less than 20% of the people express dissatisfaction.When PD=20%,RL of CO2 and HCHO are 728×10-6 and 0.068×10-6 respectively,which are stricter than the limits in the existing standards.The method proposed in this paper is applicable to 13.1%≤PD≤86.7%.展开更多
知识图谱推荐作为一种信息过滤方法被广泛应用于电子商务和网络社交等领域,然而多数基于知识图谱的推荐方法未采取合适的策略来解决传播过程中实体语义关联性衰减问题,且单维度建模无法利用知识图谱同时丰富用户和项目表示。针对以上问...知识图谱推荐作为一种信息过滤方法被广泛应用于电子商务和网络社交等领域,然而多数基于知识图谱的推荐方法未采取合适的策略来解决传播过程中实体语义关联性衰减问题,且单维度建模无法利用知识图谱同时丰富用户和项目表示。针对以上问题提出一种基于动态兴趣传播和知识图谱的推荐方法(recommendation method based on dynamic interest propagation and knowledge graph,RDPKG)。首先,通过传播网络挖掘层级用户兴趣生成用户表示,并采用注意力机制区分不同传播层数下用户兴趣的重要性;然后,通过交叉压缩单元提取知识图谱中的有效信息生成项目表示,并采用多任务学习优化推荐单元和知识图谱嵌入单元;最后,将最终的用户表示和项目表示内积获得交互概率。在推荐系统领域的3种公共数据集上进行对比实验,实验结果表明在点击率预测任务中RDPKG的准确率分别达到85.42%、76.09%和69.39%,优于其他对比方法,充分验证了RDPKG方法的有效性。展开更多
基金supporting by grant fund under the Strategic Scholarships for Frontier Research Network for the PhD Program Thai Doctoral degree
文摘A new recommendation method was presented based on memetic algorithm-based clustering. The proposed method was tested on four highly sparse real-world datasets. Its recommendation performance is evaluated and compared with that of the frequency-based, user-based, item-based, k-means clustering-based, and genetic algorithm-based methods in terms of precision, recall, and F1 score. The results show that the proposed method yields better performance under the new user cold-start problem when each of new active users selects only one or two items into the basket. The average F1 scores on all four datasets are improved by 225.0%, 61.6%, 54.6%, 49.3%, 28.8%, and 6.3% over the frequency-based, user-based, item-based, k-means clustering-based, and two genetic algorithm-based methods, respectively.
文摘From the view of both objective and subjective factors,the indoor air quality(IAQ)evaluation was considered.Carbon dioxide(CO2)and formaldehyde(HCHO)were selected as the typical contaminants of indoor air,and the evaluation method of logarithmic index was adopted as the evaluation means of IAQ.Then the recommended limits(RL)of typical contaminants CO2 and HCHO were given through analysis and calculation.The limits of CO2 and HCHO in Indoor Air Quality Standard of China or other existing standards probably correspond to the level of PD=25(%).The result shows that the existing standards fail to meet the requirement of the definition of "acceptable indoor air quality",that is to say,less than 20% of the people express dissatisfaction.When PD=20%,RL of CO2 and HCHO are 728×10-6 and 0.068×10-6 respectively,which are stricter than the limits in the existing standards.The method proposed in this paper is applicable to 13.1%≤PD≤86.7%.
文摘知识图谱推荐作为一种信息过滤方法被广泛应用于电子商务和网络社交等领域,然而多数基于知识图谱的推荐方法未采取合适的策略来解决传播过程中实体语义关联性衰减问题,且单维度建模无法利用知识图谱同时丰富用户和项目表示。针对以上问题提出一种基于动态兴趣传播和知识图谱的推荐方法(recommendation method based on dynamic interest propagation and knowledge graph,RDPKG)。首先,通过传播网络挖掘层级用户兴趣生成用户表示,并采用注意力机制区分不同传播层数下用户兴趣的重要性;然后,通过交叉压缩单元提取知识图谱中的有效信息生成项目表示,并采用多任务学习优化推荐单元和知识图谱嵌入单元;最后,将最终的用户表示和项目表示内积获得交互概率。在推荐系统领域的3种公共数据集上进行对比实验,实验结果表明在点击率预测任务中RDPKG的准确率分别达到85.42%、76.09%和69.39%,优于其他对比方法,充分验证了RDPKG方法的有效性。
