Location based social networks( LBSNs) provide location specific data generated from smart phone into online social networks thus people can share their points of interest( POIs). POI collections are complex and c...Location based social networks( LBSNs) provide location specific data generated from smart phone into online social networks thus people can share their points of interest( POIs). POI collections are complex and can be influenced by various factors,such as user preferences,social relationships and geographical influence. Therefore,recommending new locations in LBSNs requires to take all these factors into consideration. However,one problem is how to determine optimal weights of influencing factors in an algorithm in which these factors are combined. The user similarity can be obtained from the user check-in data,or from the user friend information,or based on the different geographical influences on each user's check-in activities. In this paper,we propose an algorithm that calculates the user similarity based on check-in records and social relationships,using a proposed weighting function to adjust the weights of these two kinds of similarities based on the geographical distance between users. In addition,a non-parametric density estimation method is applied to predict the unique geographical influence on each user by getting the density probability plot of the distance between every pair of user's check-in locations. Experimental results,using foursquare datasets,have shown that comparisons between the proposed algorithm and the other five baseline recommendation algorithms in LBSNs demonstrate that our proposed algorithm is superior in accuracy and recall,furthermore solving the sparsity problem.展开更多
Internet takes a role as a place for communication between people beyond a space simply for the acquisition of information.Recently,social network service(SNS)reflecting human’s basic desire for talking and communica...Internet takes a role as a place for communication between people beyond a space simply for the acquisition of information.Recently,social network service(SNS)reflecting human’s basic desire for talking and communicating with others is focused on around the world.And location-based service(LBS)is a service that provides various life conveniences like improving productivity through location information,such as GPS and WiFi.This paper suggests an application combining LBS and SNS based on Android OS.By using smart phone which is personal mobile information equipment,it combines location information with user information and SNS so that the service can be developed.It also maximizes sharing and use of information via twit based on locations of friends.This proposed system is aims for users to show online identity more actively and more conveniently.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,LBSN)中的一项重要个性化服务.由于LBSN中数据的极度稀疏性,基于协同过滤的算法推荐精度不高,文中提出基于元路径的兴趣点推荐算法.首先根据LBS...兴趣点(Point-Of-Interest,POI)推荐是基于位置的社交网络(Location-Based Social Networks,LBSN)中的一项重要个性化服务.由于LBSN中数据的极度稀疏性,基于协同过滤的算法推荐精度不高,文中提出基于元路径的兴趣点推荐算法.首先根据LBSN结构特征构建带权异构网络模型,其次引入元路径来描述节点间不同类型关联关系,基于三度影响力设置用户-兴趣点间元路径特征集,然后通过随机游走方法计算元路径特征值以度量实例路径中的首尾节点间关联度,并利用监督学习方法获得各特征的权值,最后计算特定用户将来在各兴趣点的签到概率从而生成推荐列表.文中在3个真实LBSN签到数据集上进行了实验,结果表明该算法可以有效缓解LBSN中的极度稀疏性问题,比传统推荐算法有更好的推荐效果.展开更多
如何发现高质量的社区结构对于深刻研究和分析基于位置的社交网络(location-based social networks,简称LBSN)这种新型复杂网络具有重要意义,然而,现有的面向社交网络的社区发现方法都无法适用于具有多维异构关系的LBSN.为此,提出了一...如何发现高质量的社区结构对于深刻研究和分析基于位置的社交网络(location-based social networks,简称LBSN)这种新型复杂网络具有重要意义,然而,现有的面向社交网络的社区发现方法都无法适用于具有多维异构关系的LBSN.为此,提出了一种基于联合聚类的用户社区发现方法Multi-BVD,该方法首先给出了融合用户社交网络与地理位置标签网络中多模实体及其异构关系的社区划分目标函数,然后使用拉格朗日乘子法得到目标函数极小值的迭代更新规则,并运用块值矩阵分解技术来确定最优的社区划分结果.仿真实验结果表明,Multi-BVD方法能够有效地发现LBSN中具有地理特征的用户社区结构,该社区结构在社交关系和地理兴趣标签上都有更优的内聚性,并能更紧密地体现用户社区与地理标签簇间的兴趣关联性.展开更多
近年来,异质网络中的社区发现逐渐成为人们关注的研究热点,然而现有大多数非重叠或重叠的社区发现方法都局限于考虑单一类型的网络结构,而无法适用于包含多模实体及其多维关系的异质网络,基于位置的社交网络(location based social netw...近年来,异质网络中的社区发现逐渐成为人们关注的研究热点,然而现有大多数非重叠或重叠的社区发现方法都局限于考虑单一类型的网络结构,而无法适用于包含多模实体及其多维关系的异质网络,基于位置的社交网络(location based social network,LBSN)作为最近兴起的一种新型异质网络,如何有效发现其含有多维关系的复杂社区结构对现有研究来说是一个挑战性的难题.为此,提出了一种融合用户与位置实体及其多维关系的社区发现方法MRNMF(multi-relational nonnegative matrix factorization),该方法通过建立基于非负矩阵分解的联合聚类目标函数,并考虑融入用户社交关系、用户位置签到关系以及兴趣点特征等多维度的影响因素,能同时获得紧密关联的用户模糊社区与兴趣点聚簇结构,以有效缓解推荐中的数据稀疏问题.在2种真实LBSN数据集上的实验结果表明,所提出的MRNMF方法同时在兴趣点与朋友这双重推荐上比其他传统方法具有更优越的推荐性能.展开更多
针对位置社交网络(location-based social networks,LBSN)中连续兴趣点(point-of-interest,POI)推荐系统面临的数据稀疏性、签到数据的隐式反馈属性、用户的个性化偏好等挑战,提出一种融合时空信息的连续兴趣点推荐算法。该算法将用户...针对位置社交网络(location-based social networks,LBSN)中连续兴趣点(point-of-interest,POI)推荐系统面临的数据稀疏性、签到数据的隐式反馈属性、用户的个性化偏好等挑战,提出一种融合时空信息的连续兴趣点推荐算法。该算法将用户的签到行为建模为用户-当前兴趣点-下一个兴趣点-时间段的四阶张量,并利用LBSN中的地理信息定义用户访问兴趣点的地理距离偏好,最后采用BPR(Bayesian personalized ranking)标准优化目标函数。实验结果表明该算法相比其他先进的连续兴趣点推荐算法具有更好的推荐效果。展开更多
好友推荐是基于位置的社交网络LBSN(Location-Based Social Networks)的重要服务之一。融合线上关系和线下行为,考虑位置偏好相似性、距离相似性和熟识度三个特征,构建新的好友推荐算法。通过考虑时间因素和排除时间因素两方面计算位置...好友推荐是基于位置的社交网络LBSN(Location-Based Social Networks)的重要服务之一。融合线上关系和线下行为,考虑位置偏好相似性、距离相似性和熟识度三个特征,构建新的好友推荐算法。通过考虑时间因素和排除时间因素两方面计算位置偏好的相似性;通过探究用户与其好友间签到地点在距离上的关系计算距离相似性;使用阶数与路数作为影响好友关系的重要因素计算熟识度;对以上三个特征进行加权并融合用户影响力计算最终推荐分数。利用Gowalla上的数据证明该算法可以有效提高好友推荐的有效性。展开更多
在地点推荐应用中,传统的协同过滤推荐算法由于签到数据稀疏导致推荐效果不佳。为提高推荐效果并克服传统协同过滤推荐算法受到热门地点影响的不足,提出一种新的地点推荐算法。将签到地点转换为向量,通过向量的余弦相似性计算签到地点...在地点推荐应用中,传统的协同过滤推荐算法由于签到数据稀疏导致推荐效果不佳。为提高推荐效果并克服传统协同过滤推荐算法受到热门地点影响的不足,提出一种新的地点推荐算法。将签到地点转换为向量,通过向量的余弦相似性计算签到地点的地点相似性。标记签到频次较低的地点为冷门地点,以计算签到地点的用户相似性,结合地理因素的影响,生成对用户的推荐列表。实验结果表明,相比传统协同过滤推荐算法,该算法 F 1值提升了0.009以上,推荐效果更好。展开更多
随着移动社交平台的发展,基于位置的社交网络服务(Location-Based Social Network,LBSN)已进入人们的视野。在LBSN中,根据用户的签到数据进行兴趣点(Point-of-Interest,POI)推荐是近年来研究的热点问题。提出一种基于极限学习机(Extreme...随着移动社交平台的发展,基于位置的社交网络服务(Location-Based Social Network,LBSN)已进入人们的视野。在LBSN中,根据用户的签到数据进行兴趣点(Point-of-Interest,POI)推荐是近年来研究的热点问题。提出一种基于极限学习机(Extreme Learning Machine,ELM)的POI推荐算法,提取用户的个人偏好、朋友偏好、类型偏好、流行度偏好等特征,利用ELM提供的分类方法,使用上述特征向量集合训练ELM分类器,最终根据分类结果向用户推荐POI。本文使用Foursquare和Twitter数据集的实验结果表明,该方法在精确率和效率方面均有所提高。展开更多
基于位置的社交网络(Location Based Social Networks,LBSN)的相关服务推荐越来越多,而兴趣点(Point Of Interest,POI)推荐作为LBSN相关服务中的一项个性化推荐也备受关注,越来越多的学者投入研究。目前,各种基于位置的推荐算法层出不穷...基于位置的社交网络(Location Based Social Networks,LBSN)的相关服务推荐越来越多,而兴趣点(Point Of Interest,POI)推荐作为LBSN相关服务中的一项个性化推荐也备受关注,越来越多的学者投入研究。目前,各种基于位置的推荐算法层出不穷,但由于LBSN中的数据极度稀疏的原因,导致许多算法推荐精度不高,本文提出了一种基于用户活动区域划分的元路径推荐算法。首先,根据用户签到以及点评的地点呈现区域性,将用户活动区域分为频繁活动区域和不经常活动区域,根据LBSN结构特征构建用户-活动区域和活动区域-兴趣点之间的二分图模型,其次引入元路径,计算从用户到兴趣点的实例路径的关联度,最后根据关联度大小生成推荐列表。结果表明,该算法较传统的LBSN推荐算法有更好的推荐效果。展开更多
提出一种基于位置的社交网络中利用深度学习的POI推荐方法。建立一个地理时空注意力网络,以发现总体序列依赖性和微妙的POI-POI关系;将签到序列中连续的地理距离和时间间隔信息加入到地理时空注意力网络中,建立用户个性化移动行为和挖...提出一种基于位置的社交网络中利用深度学习的POI推荐方法。建立一个地理时空注意力网络,以发现总体序列依赖性和微妙的POI-POI关系;将签到序列中连续的地理距离和时间间隔信息加入到地理时空注意力网络中,建立用户个性化移动行为和挖掘用户个性化时空偏好;设计特定于上下文的共同注意力网络,通过从签到历史中自适应选择相关签到活动来学习更改用户偏好,使地理-时空门控循环单元网络(geographical-spatiotemporal gated recurrent unit network,GS-GRUN)能够区分不同签到的用户偏好程度。在Foursquare和Gowalla数据集上的实验结果表明,所提算法能够显著提升POI推荐方法的推荐匹配度。展开更多
Recently, as location-based social network(LBSN) rapidly grow, point-of-interest(POI) recommendation has become an important way to help people locate interesting places. Nowadays, there have been deep studies conduct...Recently, as location-based social network(LBSN) rapidly grow, point-of-interest(POI) recommendation has become an important way to help people locate interesting places. Nowadays, there have been deep studies conducted on the geographical and social influence in the point-of-interest recommendation model based on the rating prediction. The fact is, however, relying solely on the rating fails to reflect the user's preferences very accurately, because the users are most concerned with the list of ranked point-of-interests(POIs) on the actual output of recommender systems. In this paper, we propose a co-pairwise ranking model called Geo-Social Bayesian Personalized Ranking model(GSBPR), which is based on the pairwise ranking with the exploiting geo-social correlations by incorporating the method of ranking learning into the process of POI recommendation. In this model, we develop a novel BPR pairwise ranking assumption by injecting users' geo-social preference. Based on this assumption, the POI recommendation model is reformulated by a three-level joint pairwise ranking model. And the experimental results based on real datasets show that the proposed method in this paper enjoys better recommendation performance compared to other state-of-the-art POI recommendation models.展开更多
Location-based social network (LBSN) is at the forefront of emerging trends in social network services (SNS) since the users in LBSN are allowed to "check-in" the places (locations) when they visit them. The a...Location-based social network (LBSN) is at the forefront of emerging trends in social network services (SNS) since the users in LBSN are allowed to "check-in" the places (locations) when they visit them. The accurate geographi- cal and temporal information of these check-in actions are provided by the end-user GPS-enabled mobile devices, and recorded by the LBSN system. In this paper, we analyze and mine a big LBSN data, Gowalla, collected by us. First, we investigate the relationship between the spatio-temporal co- occurrences and social ties, and the results show that the co- occurrences are strongly correlative with the social ties. Sec- ond, we present a study of predicting two users whether or not they will meet (co-occur) at a place in a given future time, by exploring their check-in habits. In particular, we first intro- duce two new concepts, bag-of-location and bag-of-time-lag, to characterize user's check-in habits. Based on such bag rep- resentations, we define a similarity metric called habits sim- ilarity to measure the similarity between two users' check-in habits. Then we propose a machine !earning formula for pre- dicting co-occurrence based on the social ties and habits sim- ilarities. Finally, we conduct extensive experiments on our dataset, and the results demonstrate the effectiveness of the proposed method.展开更多
文摘Location based social networks( LBSNs) provide location specific data generated from smart phone into online social networks thus people can share their points of interest( POIs). POI collections are complex and can be influenced by various factors,such as user preferences,social relationships and geographical influence. Therefore,recommending new locations in LBSNs requires to take all these factors into consideration. However,one problem is how to determine optimal weights of influencing factors in an algorithm in which these factors are combined. The user similarity can be obtained from the user check-in data,or from the user friend information,or based on the different geographical influences on each user's check-in activities. In this paper,we propose an algorithm that calculates the user similarity based on check-in records and social relationships,using a proposed weighting function to adjust the weights of these two kinds of similarities based on the geographical distance between users. In addition,a non-parametric density estimation method is applied to predict the unique geographical influence on each user by getting the density probability plot of the distance between every pair of user's check-in locations. Experimental results,using foursquare datasets,have shown that comparisons between the proposed algorithm and the other five baseline recommendation algorithms in LBSNs demonstrate that our proposed algorithm is superior in accuracy and recall,furthermore solving the sparsity problem.
基金MKE(the Ministry of Knowledge Economy),Korea,under the ITRC(Infor mation Technology Research Center)support program supervised by the NIPA(National IT Industry Promotion Agency)(NIPA-2011-C1090-1121-0010)
文摘Internet takes a role as a place for communication between people beyond a space simply for the acquisition of information.Recently,social network service(SNS)reflecting human’s basic desire for talking and communicating with others is focused on around the world.And location-based service(LBS)is a service that provides various life conveniences like improving productivity through location information,such as GPS and WiFi.This paper suggests an application combining LBS and SNS based on Android OS.By using smart phone which is personal mobile information equipment,it combines location information with user information and SNS so that the service can be developed.It also maximizes sharing and use of information via twit based on locations of friends.This proposed system is aims for users to show online identity more actively and more conveniently.
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,LBSN)中的一项重要个性化服务.由于LBSN中数据的极度稀疏性,基于协同过滤的算法推荐精度不高,文中提出基于元路径的兴趣点推荐算法.首先根据LBSN结构特征构建带权异构网络模型,其次引入元路径来描述节点间不同类型关联关系,基于三度影响力设置用户-兴趣点间元路径特征集,然后通过随机游走方法计算元路径特征值以度量实例路径中的首尾节点间关联度,并利用监督学习方法获得各特征的权值,最后计算特定用户将来在各兴趣点的签到概率从而生成推荐列表.文中在3个真实LBSN签到数据集上进行了实验,结果表明该算法可以有效缓解LBSN中的极度稀疏性问题,比传统推荐算法有更好的推荐效果.
文摘如何发现高质量的社区结构对于深刻研究和分析基于位置的社交网络(location-based social networks,简称LBSN)这种新型复杂网络具有重要意义,然而,现有的面向社交网络的社区发现方法都无法适用于具有多维异构关系的LBSN.为此,提出了一种基于联合聚类的用户社区发现方法Multi-BVD,该方法首先给出了融合用户社交网络与地理位置标签网络中多模实体及其异构关系的社区划分目标函数,然后使用拉格朗日乘子法得到目标函数极小值的迭代更新规则,并运用块值矩阵分解技术来确定最优的社区划分结果.仿真实验结果表明,Multi-BVD方法能够有效地发现LBSN中具有地理特征的用户社区结构,该社区结构在社交关系和地理兴趣标签上都有更优的内聚性,并能更紧密地体现用户社区与地理标签簇间的兴趣关联性.
文摘近年来,异质网络中的社区发现逐渐成为人们关注的研究热点,然而现有大多数非重叠或重叠的社区发现方法都局限于考虑单一类型的网络结构,而无法适用于包含多模实体及其多维关系的异质网络,基于位置的社交网络(location based social network,LBSN)作为最近兴起的一种新型异质网络,如何有效发现其含有多维关系的复杂社区结构对现有研究来说是一个挑战性的难题.为此,提出了一种融合用户与位置实体及其多维关系的社区发现方法MRNMF(multi-relational nonnegative matrix factorization),该方法通过建立基于非负矩阵分解的联合聚类目标函数,并考虑融入用户社交关系、用户位置签到关系以及兴趣点特征等多维度的影响因素,能同时获得紧密关联的用户模糊社区与兴趣点聚簇结构,以有效缓解推荐中的数据稀疏问题.在2种真实LBSN数据集上的实验结果表明,所提出的MRNMF方法同时在兴趣点与朋友这双重推荐上比其他传统方法具有更优越的推荐性能.
文摘针对位置社交网络(location-based social networks,LBSN)中连续兴趣点(point-of-interest,POI)推荐系统面临的数据稀疏性、签到数据的隐式反馈属性、用户的个性化偏好等挑战,提出一种融合时空信息的连续兴趣点推荐算法。该算法将用户的签到行为建模为用户-当前兴趣点-下一个兴趣点-时间段的四阶张量,并利用LBSN中的地理信息定义用户访问兴趣点的地理距离偏好,最后采用BPR(Bayesian personalized ranking)标准优化目标函数。实验结果表明该算法相比其他先进的连续兴趣点推荐算法具有更好的推荐效果。
文摘好友推荐是基于位置的社交网络LBSN(Location-Based Social Networks)的重要服务之一。融合线上关系和线下行为,考虑位置偏好相似性、距离相似性和熟识度三个特征,构建新的好友推荐算法。通过考虑时间因素和排除时间因素两方面计算位置偏好的相似性;通过探究用户与其好友间签到地点在距离上的关系计算距离相似性;使用阶数与路数作为影响好友关系的重要因素计算熟识度;对以上三个特征进行加权并融合用户影响力计算最终推荐分数。利用Gowalla上的数据证明该算法可以有效提高好友推荐的有效性。
文摘在地点推荐应用中,传统的协同过滤推荐算法由于签到数据稀疏导致推荐效果不佳。为提高推荐效果并克服传统协同过滤推荐算法受到热门地点影响的不足,提出一种新的地点推荐算法。将签到地点转换为向量,通过向量的余弦相似性计算签到地点的地点相似性。标记签到频次较低的地点为冷门地点,以计算签到地点的用户相似性,结合地理因素的影响,生成对用户的推荐列表。实验结果表明,相比传统协同过滤推荐算法,该算法 F 1值提升了0.009以上,推荐效果更好。
文摘随着移动社交平台的发展,基于位置的社交网络服务(Location-Based Social Network,LBSN)已进入人们的视野。在LBSN中,根据用户的签到数据进行兴趣点(Point-of-Interest,POI)推荐是近年来研究的热点问题。提出一种基于极限学习机(Extreme Learning Machine,ELM)的POI推荐算法,提取用户的个人偏好、朋友偏好、类型偏好、流行度偏好等特征,利用ELM提供的分类方法,使用上述特征向量集合训练ELM分类器,最终根据分类结果向用户推荐POI。本文使用Foursquare和Twitter数据集的实验结果表明,该方法在精确率和效率方面均有所提高。
文摘基于位置的社交网络(Location Based Social Networks,LBSN)的相关服务推荐越来越多,而兴趣点(Point Of Interest,POI)推荐作为LBSN相关服务中的一项个性化推荐也备受关注,越来越多的学者投入研究。目前,各种基于位置的推荐算法层出不穷,但由于LBSN中的数据极度稀疏的原因,导致许多算法推荐精度不高,本文提出了一种基于用户活动区域划分的元路径推荐算法。首先,根据用户签到以及点评的地点呈现区域性,将用户活动区域分为频繁活动区域和不经常活动区域,根据LBSN结构特征构建用户-活动区域和活动区域-兴趣点之间的二分图模型,其次引入元路径,计算从用户到兴趣点的实例路径的关联度,最后根据关联度大小生成推荐列表。结果表明,该算法较传统的LBSN推荐算法有更好的推荐效果。
文摘提出一种基于位置的社交网络中利用深度学习的POI推荐方法。建立一个地理时空注意力网络,以发现总体序列依赖性和微妙的POI-POI关系;将签到序列中连续的地理距离和时间间隔信息加入到地理时空注意力网络中,建立用户个性化移动行为和挖掘用户个性化时空偏好;设计特定于上下文的共同注意力网络,通过从签到历史中自适应选择相关签到活动来学习更改用户偏好,使地理-时空门控循环单元网络(geographical-spatiotemporal gated recurrent unit network,GS-GRUN)能够区分不同签到的用户偏好程度。在Foursquare和Gowalla数据集上的实验结果表明,所提算法能够显著提升POI推荐方法的推荐匹配度。
基金supported by National Basic Research Program of China (2012CB719905)National Natural Science Funds of China (41201404)Fundamental Research Funds for the Central Universities of China (2042018gf0008)
文摘Recently, as location-based social network(LBSN) rapidly grow, point-of-interest(POI) recommendation has become an important way to help people locate interesting places. Nowadays, there have been deep studies conducted on the geographical and social influence in the point-of-interest recommendation model based on the rating prediction. The fact is, however, relying solely on the rating fails to reflect the user's preferences very accurately, because the users are most concerned with the list of ranked point-of-interests(POIs) on the actual output of recommender systems. In this paper, we propose a co-pairwise ranking model called Geo-Social Bayesian Personalized Ranking model(GSBPR), which is based on the pairwise ranking with the exploiting geo-social correlations by incorporating the method of ranking learning into the process of POI recommendation. In this model, we develop a novel BPR pairwise ranking assumption by injecting users' geo-social preference. Based on this assumption, the POI recommendation model is reformulated by a three-level joint pairwise ranking model. And the experimental results based on real datasets show that the proposed method in this paper enjoys better recommendation performance compared to other state-of-the-art POI recommendation models.
文摘Location-based social network (LBSN) is at the forefront of emerging trends in social network services (SNS) since the users in LBSN are allowed to "check-in" the places (locations) when they visit them. The accurate geographi- cal and temporal information of these check-in actions are provided by the end-user GPS-enabled mobile devices, and recorded by the LBSN system. In this paper, we analyze and mine a big LBSN data, Gowalla, collected by us. First, we investigate the relationship between the spatio-temporal co- occurrences and social ties, and the results show that the co- occurrences are strongly correlative with the social ties. Sec- ond, we present a study of predicting two users whether or not they will meet (co-occur) at a place in a given future time, by exploring their check-in habits. In particular, we first intro- duce two new concepts, bag-of-location and bag-of-time-lag, to characterize user's check-in habits. Based on such bag rep- resentations, we define a similarity metric called habits sim- ilarity to measure the similarity between two users' check-in habits. Then we propose a machine !earning formula for pre- dicting co-occurrence based on the social ties and habits sim- ilarities. Finally, we conduct extensive experiments on our dataset, and the results demonstrate the effectiveness of the proposed method.