By mining of the requirements of lots of electric vehicle users for charging piles, this paper proposes the charging pile siting algorithm via the fusion of Points of Interest and vehicle trajectories. The proposed al...By mining of the requirements of lots of electric vehicle users for charging piles, this paper proposes the charging pile siting algorithm via the fusion of Points of Interest and vehicle trajectories. The proposed algorithm computes appropriate charging pile locations by: 1) mining user Points of Interest from social network; 2) mining parking sites of vehicle form GPS trajectories and 3) fusing the Points of Interest and parking sites together then clustering the fusions with our improved DBSCAN algorithm, whose clustering results indicates the final appropriate charging pile locations. Experimental results show that our proposed methods are more efficient than existing methods.展开更多
Cruising route recommendation based on trajectory mining can improve taxi-drivers'income and reduce energy consumption.However,existing methods mostly recommend pick-up points for taxis only.Moreover,their perform...Cruising route recommendation based on trajectory mining can improve taxi-drivers'income and reduce energy consumption.However,existing methods mostly recommend pick-up points for taxis only.Moreover,their performance is not good enough since there lacks a good evaluation model for the pick-up points.Therefore,we propose an entropy-based model for recommendation of taxis'cruising route.Firstly,we select more positional attributes from historical pick-up points in order to obtain accurate spatial-temporal features.Secondly,the information entropy of spatial-temporal features is integrated in the evaluation model.Then it is applied for getting the next pick-up points and further recommending a series of successive points.These points are constructed a cruising route for taxi-drivers.Experimental results show that our method is able to obviously improve the recommendation accuracy of pick-up points,and help taxi-drivers make profitable benefits more than before.展开更多
The rapid development of location-based social networks(LBSNs) provides people with an opportunity of better understanding their mobility behavior which enables them to decide their next location.For example,it can he...The rapid development of location-based social networks(LBSNs) provides people with an opportunity of better understanding their mobility behavior which enables them to decide their next location.For example,it can help travelers to choose where to go next,or recommend salesmen the most potential places to deliver advertisements or sell products.In this paper,a method for recommending points of interest(POIs)is proposed based on a collaborative tensor factorization(CTF)technique.Firstly,a generalized objective function is constructed for collaboratively factorizing a tensor with several feature matrices.Secondly,a 3-mode tensor is used to model all users' check-in behaviors,and three feature matrices are extracted to characterize the time distribution,category distribution and POI correlation,respectively.Thirdly,each user's preference to a POI at a specific time can be estimated by using CTF.In order to further improve the recommendation accuracy,PCTF(Partitionbased CTF) is proposed to fill the missing entries of a tensor after clustering its every mode.Experiments on a real checkin database show that the proposed method can provide more accurate location recommendation.展开更多
In this work, a best answer recommendation model is proposed for a Question Answering (QA) system. A Community Question Answering System was subsequently developed based on the model. The system applies Brouwer Fixed ...In this work, a best answer recommendation model is proposed for a Question Answering (QA) system. A Community Question Answering System was subsequently developed based on the model. The system applies Brouwer Fixed Point Theorem to prove the existence of the desired voter scoring function and Normalized Google Distance (NGD) to show closeness between words before an answer is suggested to users. Answers are ranked according to their Fixed-Point Score (FPS) for each question. Thereafter, the highest scored answer is chosen as the FPS Best Answer (BA). For each question asked by user, the system applies NGD to check if similar or related questions with the best answer had been asked and stored in the database. When similar or related questions with the best answer are not found in the database, Brouwer Fixed point is used to calculate the best answer from the pool of answers on a question then the best answer is stored in the NGD data-table for recommendation purpose. The system was implemented using PHP scripting language, MySQL for database management, JQuery, and Apache. The system was evaluated using standard metrics: Reciprocal Rank, Mean Reciprocal Rank (MRR) and Discounted Cumulative Gain (DCG). The system eliminated longer waiting time faced by askers in a community question answering system. The developed system can be used for research and learning purposes.展开更多
下一个兴趣点推荐是推荐算法领域的热点,旨在为用户推荐适合的下一地点。较新的研究通过图和序列方法模拟用户与POI的交互以及POI之间转换关系,性能得到显著提升。然而,现有模型仍然存在需要解决的问题。针对现有的下一个兴趣点推荐模...下一个兴趣点推荐是推荐算法领域的热点,旨在为用户推荐适合的下一地点。较新的研究通过图和序列方法模拟用户与POI的交互以及POI之间转换关系,性能得到显著提升。然而,现有模型仍然存在需要解决的问题。针对现有的下一个兴趣点推荐模型的局限性,特别是如何充分捕捉User-POI交互图上全局和局部信息,以及缓解图神经网络的过平滑特性导致图上信息丢失的问题,提出了基于graph Transformer的多编码模型(multi-coding network based on GT model)对下一个兴趣点进行推荐。首先,从位置和结构的视角上联合对user-POI交互图上进行全局、局部以及相对信息进行编码;然后,将编码后生成的图嵌入通过graph Transformer网络层更新图上节点与边信息;最后通过MLP网络层生成预测;最终,MCGT在Gowalla和TKY两个公开数据集进行对比实验。结果表明,在Gowalla数据集上recall和NDCG指标至少有3.79%的提升,在TKY数据集上recall和NDCG指标至少有2.5%的提升,证明了MCGT设计的合理性与有效性。展开更多
现有的大多数兴趣点(point of interest,POI)推荐系统由于忽略了用户签到序列中的顺序行为模式,以及用户的个性化偏好对于POI推荐的影响,导致POI推荐系统性能较低,推荐结果不可靠,进而影响用户体验。为了解决上述问题,提出一种融合时序...现有的大多数兴趣点(point of interest,POI)推荐系统由于忽略了用户签到序列中的顺序行为模式,以及用户的个性化偏好对于POI推荐的影响,导致POI推荐系统性能较低,推荐结果不可靠,进而影响用户体验。为了解决上述问题,提出一种融合时序门控图神经网络的兴趣点推荐方法。运用时序门控图神经网络(temporal gated graph neural network,TGGNN)学习POI embedding;采用注意力机制捕获用户的长期偏好;通过注意力机制融合用户的最新偏好和实时偏好,进而捕获用户的短期偏好。通过自适应的方式结合用户的长期和短期偏好,计算候选POI的推荐得分,并根据得分为用户进行POI推荐。实验结果表明,与现有方法相比,该方法在召回率和平均倒数排名这两项指标上均有较为明显的提升,因此可以取得很好的推荐效果,具有良好的应用前景。展开更多
兴趣点(Point-Of-Interest,POI)推荐是基于位置的社交网络(Location-based Social Networks,LBSNs)研究中最重要的任务之一。为了解决POI推荐中的空间稀疏性问题,提出一种用于位置推荐的长短期偏好时空注意力网络(LSAN)。首先,构建了签...兴趣点(Point-Of-Interest,POI)推荐是基于位置的社交网络(Location-based Social Networks,LBSNs)研究中最重要的任务之一。为了解决POI推荐中的空间稀疏性问题,提出一种用于位置推荐的长短期偏好时空注意力网络(LSAN)。首先,构建了签到序列的时空关系矩阵,使用多头注意力机制从中提取非连续签到和非相邻位置中的时空相关性,缓解签到数据稀疏所带来的困难。其次,在模型中设置用户短期偏好和长期偏好提取模块,自适应的将二者结合在一起,考虑了用户偏好对用户决策影响。最后,在Foursquare数据集上进行测试,并与其它模型结果进行对比,证实了提出的LSAN模型结果最优。研究表明LSAN模型能够获得最佳的推荐效果,为POI推荐提供新思路。展开更多
为了提高兴趣点(point of interest,POI)推荐的准确性和个性化,提升用户对推荐结果的满意度,针对不同活跃度用户的特点,提出一种融合用户活跃度的上下文感知兴趣点推荐算法(A POI recommendation algorithm that integrates geographica...为了提高兴趣点(point of interest,POI)推荐的准确性和个性化,提升用户对推荐结果的满意度,针对不同活跃度用户的特点,提出一种融合用户活跃度的上下文感知兴趣点推荐算法(A POI recommendation algorithm that integrates geographical,categorical,and temporal factors,while simultaneously considering user activity),简称AU-GCTRS。首先,为缓解数据稀疏性和冷启动问题,引入多维上下文信息;其次,通过挖掘用户签到频率、签到兴趣点数量和签到时间,将用户划分为不同活跃度的群体;最后,综合用户活跃度与上下文分数,将得分高的前K个兴趣点推荐给用户。在真实数据集上进行实验表明,AU-GCTRS算法比其他流行算法更有效地缓解了数据稀疏性和冷启动问题,提高了推荐准确率和召回率。展开更多
The wide spread of location-based social networks brings about a huge volume of user check-in data, whichfacilitates the recommendation of points of interest (POIs). Recent advances on distributed representation she...The wide spread of location-based social networks brings about a huge volume of user check-in data, whichfacilitates the recommendation of points of interest (POIs). Recent advances on distributed representation shed light onlearning low dimensional dense vectors to alleviate the data sparsity problem. Current studies on representation learningfor POI recommendation embed both users and POIs in a common latent space, and users' preference is inferred basedon the distance/similarity between a user and a POI. Such an approach is not in accordance with the semantics of usersand POIs as they are inherently different objects. In this paper, we present a novel translation-based, time and locationaware (TransTL) representation, which models the spatial and temporal information as a relationship connecting users andPOIs. Our model generalizes the recent advances in knowledge graph embedding. The basic idea is that the embedding ofa 〈time, location〉 pair corresponds to a translation from embeddings of users to POIs. Since the POI embedding shouldbe close to the user embedding plus the relationship vector, the recommendation can be performed by selecting the top-kPOIs similar to the translated POI, which are all of the same type of objects. We conduct extensive experiments on tworeal-world data.sets. The results demonstrate that our TransTL model achieves the state-of-the-art performance. It is alsomuch more robust to data sparsity than the baselines.展开更多
文摘By mining of the requirements of lots of electric vehicle users for charging piles, this paper proposes the charging pile siting algorithm via the fusion of Points of Interest and vehicle trajectories. The proposed algorithm computes appropriate charging pile locations by: 1) mining user Points of Interest from social network; 2) mining parking sites of vehicle form GPS trajectories and 3) fusing the Points of Interest and parking sites together then clustering the fusions with our improved DBSCAN algorithm, whose clustering results indicates the final appropriate charging pile locations. Experimental results show that our proposed methods are more efficient than existing methods.
基金funded by the National Natural Science Foundation of China(61872139,41871320)Provincial and Municipal Joint Fund of Hunan Provincial Natural Science Foundation of China(2018JJ4052)+2 种基金Hunan Provincial Natural Science Foundation of China(2017JJ2081)the Key Project of Hunan Provincial Education Department(17A070,19A172)the Project of Hunan Provincial Education Department(17C0646).
文摘Cruising route recommendation based on trajectory mining can improve taxi-drivers'income and reduce energy consumption.However,existing methods mostly recommend pick-up points for taxis only.Moreover,their performance is not good enough since there lacks a good evaluation model for the pick-up points.Therefore,we propose an entropy-based model for recommendation of taxis'cruising route.Firstly,we select more positional attributes from historical pick-up points in order to obtain accurate spatial-temporal features.Secondly,the information entropy of spatial-temporal features is integrated in the evaluation model.Then it is applied for getting the next pick-up points and further recommending a series of successive points.These points are constructed a cruising route for taxi-drivers.Experimental results show that our method is able to obviously improve the recommendation accuracy of pick-up points,and help taxi-drivers make profitable benefits more than before.
基金supported in part by the National Nature Science Foundation of China(91218301,61572360)the Basic Research Projects of People's Public Security University of China(2016JKF01316)Shanghai Shuguang Program(15SG18)
文摘The rapid development of location-based social networks(LBSNs) provides people with an opportunity of better understanding their mobility behavior which enables them to decide their next location.For example,it can help travelers to choose where to go next,or recommend salesmen the most potential places to deliver advertisements or sell products.In this paper,a method for recommending points of interest(POIs)is proposed based on a collaborative tensor factorization(CTF)technique.Firstly,a generalized objective function is constructed for collaboratively factorizing a tensor with several feature matrices.Secondly,a 3-mode tensor is used to model all users' check-in behaviors,and three feature matrices are extracted to characterize the time distribution,category distribution and POI correlation,respectively.Thirdly,each user's preference to a POI at a specific time can be estimated by using CTF.In order to further improve the recommendation accuracy,PCTF(Partitionbased CTF) is proposed to fill the missing entries of a tensor after clustering its every mode.Experiments on a real checkin database show that the proposed method can provide more accurate location recommendation.
文摘In this work, a best answer recommendation model is proposed for a Question Answering (QA) system. A Community Question Answering System was subsequently developed based on the model. The system applies Brouwer Fixed Point Theorem to prove the existence of the desired voter scoring function and Normalized Google Distance (NGD) to show closeness between words before an answer is suggested to users. Answers are ranked according to their Fixed-Point Score (FPS) for each question. Thereafter, the highest scored answer is chosen as the FPS Best Answer (BA). For each question asked by user, the system applies NGD to check if similar or related questions with the best answer had been asked and stored in the database. When similar or related questions with the best answer are not found in the database, Brouwer Fixed point is used to calculate the best answer from the pool of answers on a question then the best answer is stored in the NGD data-table for recommendation purpose. The system was implemented using PHP scripting language, MySQL for database management, JQuery, and Apache. The system was evaluated using standard metrics: Reciprocal Rank, Mean Reciprocal Rank (MRR) and Discounted Cumulative Gain (DCG). The system eliminated longer waiting time faced by askers in a community question answering system. The developed system can be used for research and learning purposes.
文摘下一个兴趣点推荐是推荐算法领域的热点,旨在为用户推荐适合的下一地点。较新的研究通过图和序列方法模拟用户与POI的交互以及POI之间转换关系,性能得到显著提升。然而,现有模型仍然存在需要解决的问题。针对现有的下一个兴趣点推荐模型的局限性,特别是如何充分捕捉User-POI交互图上全局和局部信息,以及缓解图神经网络的过平滑特性导致图上信息丢失的问题,提出了基于graph Transformer的多编码模型(multi-coding network based on GT model)对下一个兴趣点进行推荐。首先,从位置和结构的视角上联合对user-POI交互图上进行全局、局部以及相对信息进行编码;然后,将编码后生成的图嵌入通过graph Transformer网络层更新图上节点与边信息;最后通过MLP网络层生成预测;最终,MCGT在Gowalla和TKY两个公开数据集进行对比实验。结果表明,在Gowalla数据集上recall和NDCG指标至少有3.79%的提升,在TKY数据集上recall和NDCG指标至少有2.5%的提升,证明了MCGT设计的合理性与有效性。
文摘兴趣点(Point-Of-Interest,POI)推荐是基于位置的社交网络(Location-based Social Networks,LBSNs)研究中最重要的任务之一。为了解决POI推荐中的空间稀疏性问题,提出一种用于位置推荐的长短期偏好时空注意力网络(LSAN)。首先,构建了签到序列的时空关系矩阵,使用多头注意力机制从中提取非连续签到和非相邻位置中的时空相关性,缓解签到数据稀疏所带来的困难。其次,在模型中设置用户短期偏好和长期偏好提取模块,自适应的将二者结合在一起,考虑了用户偏好对用户决策影响。最后,在Foursquare数据集上进行测试,并与其它模型结果进行对比,证实了提出的LSAN模型结果最优。研究表明LSAN模型能够获得最佳的推荐效果,为POI推荐提供新思路。
文摘为了提高兴趣点(point of interest,POI)推荐的准确性和个性化,提升用户对推荐结果的满意度,针对不同活跃度用户的特点,提出一种融合用户活跃度的上下文感知兴趣点推荐算法(A POI recommendation algorithm that integrates geographical,categorical,and temporal factors,while simultaneously considering user activity),简称AU-GCTRS。首先,为缓解数据稀疏性和冷启动问题,引入多维上下文信息;其次,通过挖掘用户签到频率、签到兴趣点数量和签到时间,将用户划分为不同活跃度的群体;最后,综合用户活跃度与上下文分数,将得分高的前K个兴趣点推荐给用户。在真实数据集上进行实验表明,AU-GCTRS算法比其他流行算法更有效地缓解了数据稀疏性和冷启动问题,提高了推荐准确率和召回率。
基金This work was supported by the National Natural Science Foundation of China under Grant Nos. 61572376 and 91646206, and the National Key Research and Development Program of China under Grant No. 2016YFB1000603.
文摘The wide spread of location-based social networks brings about a huge volume of user check-in data, whichfacilitates the recommendation of points of interest (POIs). Recent advances on distributed representation shed light onlearning low dimensional dense vectors to alleviate the data sparsity problem. Current studies on representation learningfor POI recommendation embed both users and POIs in a common latent space, and users' preference is inferred basedon the distance/similarity between a user and a POI. Such an approach is not in accordance with the semantics of usersand POIs as they are inherently different objects. In this paper, we present a novel translation-based, time and locationaware (TransTL) representation, which models the spatial and temporal information as a relationship connecting users andPOIs. Our model generalizes the recent advances in knowledge graph embedding. The basic idea is that the embedding ofa 〈time, location〉 pair corresponds to a translation from embeddings of users to POIs. Since the POI embedding shouldbe close to the user embedding plus the relationship vector, the recommendation can be performed by selecting the top-kPOIs similar to the translated POI, which are all of the same type of objects. We conduct extensive experiments on tworeal-world data.sets. The results demonstrate that our TransTL model achieves the state-of-the-art performance. It is alsomuch more robust to data sparsity than the baselines.