首页 | 本学科首页   官方微博 | 高级检索  
文章检索
  按 检索   检索词:      
出版年份:   被引次数:   他引次数: 提示:输入*表示无穷大
  收费全文   38篇
  免费   5篇
测绘学   7篇
大气科学   5篇
地球物理   2篇
地质学   12篇
海洋学   4篇
综合类   6篇
自然地理   7篇
  2023年   1篇
  2022年   4篇
  2021年   4篇
  2020年   2篇
  2019年   7篇
  2018年   4篇
  2017年   1篇
  2016年   1篇
  2014年   3篇
  2013年   2篇
  2011年   3篇
  2008年   3篇
  2007年   3篇
  2006年   1篇
  2005年   2篇
  2004年   1篇
  2002年   1篇
排序方式: 共有43条查询结果,搜索用时 15 毫秒
1.
建立适用于专题地图产品检索的用户偏好推荐模型是提高专题地图质量的有效方式之一,在专题地图产品推荐场景中,存在严重的内容冷启动和评论数据稀疏问题,现有的推荐算法无法为特定类用户推荐不同特征的专题地图产品,导致用户从专题地图中获取偏好信息受到限制。因此,本文构建基于负采样的连续词袋模型和基于Word2Vec的Item2Vec相结合的用户偏好推荐方法,用于专题地图产品推荐。① 计算用户行为日志文件中交互行为数据的隐性评分,以代替专题地图推荐信息流场景中稀疏的用户评论数据;② 基于负采样的连续词袋模型提取目标专题地图的前后地图序列感知特征信息,通过控制正负样本比例为1:2,提升目标专题地图潜在评分的预测精度;③ 通过Item2Vec将带有用户行为特征信息的专题地图映射到向量空间,计算用户对专题地图的相似度矩阵,根据用户偏好程度完成推荐。在构建的专题地图评分实验数据集Thematic CMaps和4个公开验证数据集MovieLens上的测试结果表明:与LFM、Personal Rank、Content Based和SVD 4种传统推荐算法相比,本文所提方法可有效提高潜在评分的预测精度,推荐性能最高达到27.85%;与以霍夫曼采样方式的Item2Vec基础方法和YouTubeNet 2种神经网络推荐算法相比,评分预测精度有一定提高,且推荐性能不断提升,最高达到2.97%和5.78%。以经典算法奇异值分解(SVD)为例,将MovieLens-20M数据集切分后,在数据量不断增大的数据子集中,本文所用方法的评分预测精度和性能均优于SVD方法。  相似文献   
2.
Matrix factorization is one of the most popular methods in recommendation systems. However, it faces two challenges related to the check-in data in point of interest (POI) recommendation: data scarcity and implicit feedback. To solve these problems, we propose a Feature-Space Separated Factorization Model (FSS-FM) in this paper. The model represents the POI feature spaces as separate slices, each of which represents a type of feature. Thus, spatial and temporal information and other contexts can be easily added to compensate for scarce data. Moreover, two commonly used objective functions for the factorization model, the weighted least squares and pairwise ranking functions, are combined to construct a hybrid optimization function. Extensive experiments are conducted on two real-life data sets: Gowalla and Foursquare, and the results are compared with those of baseline methods to evaluate the model. The results suggest that the FSS-FM performs better than state-of-the-art methods in terms of precision and recall on both data sets. The model with separate feature spaces can improve the performance of recommendation. The inclusion of spatial and temporal contexts further leverages the performance, and the spatial context is more influential than the temporal context. In addition, the capacity of hybrid optimization in improving POI recommendation is demonstrated.  相似文献   
3.
大众在旅游途中期望获得开销低、行程方便、舒适度高的旅游体验,同时还具有历史人文、自然景观、美食购物等不同游览需求.因此,本文提出了一种基于改进混合蛙跳算法的个性化旅游路线推荐方法.首先建立个性化旅游路线推荐问题的优化模型,并针对该模型的特点,设计改进混合蛙跳算法.通过调整可控精度,增加筛选准则和及时处理异常解等策略增强群体的多样性,降低遗漏最优解的风险,强化局部搜索能力,并提高算法的求解精度.以南京三日游个性化旅游路线推荐问题作为实例,收集南京市内知名景点的门票价格、开放时间、不同出行方式所需的时间和花费情况以及食宿费用等相关数据,基于改进混合蛙跳算法进行求解.实验结果表明,与改进前的方法相比,所提改进方法能够获取更优的路径解,推荐的路线能够更好地满足用户的个性需求.  相似文献   
4.
对商丹断裂带东段基性-超基性岩带(体)及其中的铁、铬等矿化特征作了概述,在岩带中圈出了A、B、C、D四个基性-超基性岩体集中区,对该带中铁铬镍等矿产资源潜力作了地质预测,认为该基性-超基性岩带是岩浆型铁、铬、镍等矿产的良好找矿远景区,提出了下步找矿工作思路、方法手段的建议.  相似文献   
5.
An investigation of girder bridges on National Highway 213 and the Doujiangyan-Wenchuan expressway after the Wenchuan earthquake showed that typical types of damage included: span collapses due to unseating at expansion joints; shear key failure; and damage of the expansion joint due to the slide-induced large relative displacement between the bottom of the girder and the top of the laminated-rubber bearing. This slide, however, can actually act as a form of isolation for the substructure, and as a result, the piers and foundation of most of the bridges on state route 213 suffered minor damage. The exception was the Baihua Bridge, which suffered severe damage. Corresponding seismic design recommendations are presented based on this investigation.  相似文献   
6.
由于网络上每天有海量的新闻报道产生,新闻推荐已经成为减轻用户信息负载、实现个性化新闻信息获取的重要途径,并被广泛用于新闻网站和新闻APP中以提升用户体验.不同于传统的商品推荐,在新闻推荐中新的新闻文章产生速度很快,而且新闻的语义信息需要结合整体新闻文本去理解,给传统的基于ID和基于特征的推荐算法带来了很大的挑战.此外,用户的新闻阅读兴趣存在高度多样性和动态性的特点,使得准确的用户建模变得非常困难.本文介绍了一些基于深度学习的个性化新闻推荐算法,并探讨了新闻推荐未来的一些可行的方向.  相似文献   
7.
How to exploit various features of users and points of interest (POIs) for accurate POI recommendation is important in location-based social networks (LBSNs). In this paper, a novel POI recommendation framework, named RecNet, is proposed, which is developed based on a deep neural network (DNN) to incorporate various features in LBSNs and learn their joint influence on user behavior. More specifically, co-visiting, geographical and categorical influences in LBSNs are exploited to alleviate the data sparsity issue in POI recommendation and are converted to feature vector representations of POIs and users via feature embedding. Moreover, the embedded POIs and users are fed into a DNN pairwise to adaptively learn high-order interactions between features. Our method is evaluated on two publicly available LBSNs datasets and experimental results show that RecNet outperforms state-of-the-art algorithms for POI recommendation.  相似文献   
8.
Currently, very few roof shape information for complex buildings is available on OSM. Moreover, additional data requirements (e.g. 3D point clouds) limit the applicability of many roof reconstruction approaches. To mitigate this issue, we propose an approach to roof shape recommendations for complex buildings by exploring the inherited characteristics of building footprints: the disclosure of rectangles combinations in a partition of footprints and the symmetrical features of footprints. First, it decomposes a complex footprint into rectangles by using an advanced minimal non-overlapping cover algorithm. Second, a graph-based symmetry detection algorithm is proposed to identify all the symmetrical sub-clusters in partitions. Then, a set of selection rules are defined to rank partitions, and the best ones are chosen for roof shape recommendation. Finally, a set of combination rules and a symmetry rule are defined. It enables to evaluate the probability of a footprint being a certain combination of roof shapes. Experimental results show the growth of the probability of correctly recommending roof shapes for single rectangles and buildings from a prior probability of 17–45% and from a prior probability of 0.29–14.3%, removing 60% and 93% of the incorrect roof shape options, respectively.  相似文献   
9.
信息技术的快速发展导致信息过载.推荐系统是解决信息过载最有效的方式之一.近年来,深度学习的快速发展也带动了推荐系统的进步,各种深度推荐算法层出不穷.然而由于候选物品数量巨大且用户兴趣动态变化,深度推荐算法的推荐复杂度巨大,难以在实际系统中单独使用.在深度推荐技术发展的同时,物品召回技术(也称近似搜索技术)也有了较大的发展与进步.本文先介绍基于距离最小化的物品召回的研究进展,再从向量索引、局部敏感哈希、哈希学习、向量量化四个方面来深入探讨基于内积最大化的物品召回技术的研究进展.  相似文献   
10.
为了从大量的信息中检索出符合用户偏好和需求的旅游资源,文中提出了综合兴趣度的概念,综合兴趣度是对用户兴趣的完整描述。在此基础上给出一种基于用户综合兴趣度的推荐方法。将用户主动提供的兴趣倾向与用户的浏览行为相结合,分析其综合兴趣。依据与目标用户兴趣相似的邻居用户对景点的评分信息预测用户对待推荐景点的评分,并选择预测评分处于前N位的景点进行推荐。  相似文献   
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号