首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献，搜索用时 31 毫秒
1.
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   

2.
随着移动通信与LBS的蓬勃发展,能够描述个体行为的众源时空大数据大量涌现,为感知群体时空行为模式与探究个性化路线提供了新视角。该文将众源时空信息与出行者的个人意愿映射到实际路网空间,融合大众偏好和定制趋势,构建包含主题序列生成、POI推荐、历史路线推荐的局部路网模型,进而实现一种利用众源时空数据改进的HMM路线规划方法,为用户提供合适且个性化的出行方案;以长沙市岳麓区为研究案例,利用真实路网数据与相关兴趣点作为实验数据,基于该方法可在短时间内提供满足用户需求的不同月份的最优路线。  相似文献   

3.
ABSTRACT

The temporal nature of humans interaction with Points of Interest (POIs) in cities can differ depending on place type and regional location. Times when many people are likely to visit restaurants (place type) in Italy, may differ from times when many people are likely to visit restaurants in Lebanon (i.e. regional differences). Geosocial data are a powerful resource to model these temporal differences in cities, as traditional methods used to study cross-cultural differences do not scale to a global level. As cities continue to grow in population and economic development, research identifying the social and geophysical (e.g., climate) factors that influence city function remains important and incomplete. In this work, we take a quantitative approach, applying dynamic time warping and hierarchical clustering on temporal signatures to model geosocial temporal patterns for Retail and Restaurant Facebook POIs hours of operation for more than 100 cities in 90 countries around the world. Results show cities’ temporal patterns cluster to reflect the cultural region they represent. Furthermore, temporal patterns are influenced by a mix of social and geophysical factors. Trends in the data suggest social factors influence unique drops in temporal signatures, and geophysical factors influence when daily temporal patterns start and finish.  相似文献   

4.
We examined three different ways to integrate spatial and temporal data in kernel density estimation methods (KDE) to identify space–time clusters of geographic events. Spatial data and time data are typically measured in different units along respective dimensions. Therefore, spatial KDE methods require special extensions when incorporating temporal data to detect spatiotemporal clusters of geographical event. In addition to a real-world data set, we applied the proposed methods to simulated data that were generated through random and normal processes to compare results of different kernel functions. The comparison is based on hit rates and values of a compactness index with considerations of both spatial and temporal attributes of the data. The results show that the spatiotemporal KDE (STKDE) can reach higher hit rates while keeping identified hotspots compact. The implementation of these STKDE methods is tested using the 2012 crime event data in Akron, Ohio, as an example. The results show that STKDE methods reveal new perspectives from the data that go beyond what can be extracted by using the conventional spatial KDE.  相似文献   

5.
A Pedagogic Framework to Link GIS to the Intellectual Core of Geography   总被引：4，自引：3，他引：1  
《The Journal of geography》2012,111(6):578-591
Abstract

This paper aims to develop a new pedagogic framework for teaching GIS at the college and university level using Berry's geographic matrix. By synthesizing different schools of thought, this paper argues that GIS education essentially involves two aspects—how to teach about GIS and how to teach with GIS. Berry's geographic matrix can be used to tie these dual aspects of GIS education together neatly. As an abstract representation of geographical phenomena, the geographic matrix embeds all three entities of GIS—location, attribute, time—and thus can help GIS instructors teach about GIS. As a synthesis of geographical approaches, the geographic matrix can assist GIS instructors teach with GIS. This paper demonstrates that GIS is actually an implementation of Berry's geographic matrix. Furthermore, the 10 approaches to geographical analysis, originally proposed by Berry for the geographic matrix, can be executed routinely in a GIS environment. By incorporating Berry's geographic matrix into GIS education, teachers can enable students to surpass technical issues and to appreciate the conceptual and functional linkages between GIS and geography's intellectual core.  相似文献   

6.
With the popularity of mobile devices and smartphones, we have witnessed rapid growth in mobile applications and services, especially in location-based services (LBS). According to a mobile marketing survey, maps/location searches are among the most utilized services on smartphones. Points of interest (POIs), such as stores, shops, gas stations, parking lots, and bus stops, are particularly important for maps/location searches. Existing map services such as Google Maps and Wikimapia are constructed manually either professionally or with crowd sourcing. However, manual annotation is costly and limited in current POI search services. With the abundance of information on the Web, many store POIs can be extracted from the Web. In this paper, we focus on automatically constructing a POI database to enable store POI map searches. We propose techniques that are required to construct a POI database, including focused crawling, information extraction, and information retrieval techniques. We first crawl Yellow Page web sites to obtain vocabularies of store names. These vocabularies are then investigated with search engines to obtain sentences containing these store names from search snippets in order to train a store name recognition model. To extract POIs scattered across the Web, we propose a query-based crawler to find address-bearing pages that might be used to extract addresses and store names. We crawled 1.25 million distinct POI pairs scattered across the Web and implemented a POI search service via Apache Lucent’s search platform, called Solr. The experimental results demonstrate that the proposed geographical information retrieval model outperforms Wikimapia and a commercial app called ‘What’s the Number?’  相似文献   

7.
ABSTRACT

Though global-coverage urban perception datasets have been recently created using machine learning, their efficacy in accurately assessing local urban perceptions for other countries and regions remains a problem. Here we describe a human-machine adversarial scoring framework using a methodology that incorporates deep learning and iterative feedback with recommendation scores, which allows for the rapid and cost-effective assessment of the local urban perceptions for Chinese cities. Using the state-of-the-art Fully Convolutional Network (FCN) and Random Forest (RF) algorithms, the proposed method provides perception estimations with errors less than 10%. The driving factor analysis from both the visual and urban functional aspects demonstrated its feasibility in facilitating local urban perception derivations. With high-throughput and high-accuracy scorings, the proposed human-machine adversarial framework offers an affordable and rapid solution for urban planners and researchers to conduct local urban perception assessments.  相似文献   

8.
Geo-tagged travel photos on social networks often contain location data such as points of interest (POIs), and also users’ travel preferences. In this paper, we propose a hybrid ensemble learning method, BAyes-Knn, that predicts personalized tourist routes for travelers by mining their geographical preferences from these location-tagged data. Our method trains two types of base classifiers to jointly predict the next travel destination: (1) The K-nearest neighbor (KNN) classifier quantifies users’ location history, weather condition, temperature and seasonality and uses a feature-weighted distance model to predict a user’s personalized interests in an unvisited location. (2) A Bayes classifier introduces a smooth kernel function to estimate a-priori probabilities of features and then combines these probabilities to predict a user’s latent interests in a location. All the outcomes from these subclassifiers are merged into one final prediction result by using the Borda count voting method. We evaluated our method on geo-tagged Flickr photos and Beijing weather data collected from 1 January 2005 to 1 July 2016. The results demonstrated that our ensemble approach outperformed 12 other baseline models. In addition, the results showed that our framework has better prediction accuracy than do context-aware significant travel-sequence-patterns recommendations and frequent travel-sequence patterns.  相似文献   

9.
Abstract

A great deal of recent work in computer science applies itself to temporal database design. Designers of geographical information systems (GIS) can build on this work to create a temporal GIS capability to trace geographical change and understand geographical processes. This paper reviews temporal research in information processing, contrasts various proposed temporal database designs and summarizes the problems of adapting it to GIS requirements.  相似文献   

10.
Assessment issues in geographic education for the twenty-first century   总被引：1，自引：1，他引：0  
《The Journal of geography》2012,111(4):171-174
  相似文献   

11.
ABSTRACT

Studies on the riding modes of shared bicycles have aimed to heighten the understanding of cycling characteristics. This paper analyzes the spatiotemporal riding modes of shared bicycles based on tensor decomposition in Beijing, China. Two third-order tensors are constructed for the origin and destination points of shared bicycles in the day, hour, and space dimensions. Three factor matrices explicitly reveal two modes, three modes, and six modes in the day dimension, hour dimension, and space dimension, respectively. The relationships among the different modes in the three dimensions are demonstrated in an interaction table. Further, the density for different types of points of interest (POIs) are calculated to further analyze the potential riding purpose for different riding modes. Notably, the main POI types for the areas of O2 and D2 modes are consistent with the areas of D3 and O3 modes, which reflects the tidal characteristics of the commuting activities of shared bicycles. The main functional areas are inferred according to the riding modes and POIs, which enables verification of the correctness of the obtained riding modes to some extent. By method comparison, tensor decomposition shows the advantage of being able to reveal the spatiotemporal modes among multiple dimensions.  相似文献   

12.
Abstract

This paper presents an argument for introducing location-allocation theory to advanced undergraduate and beginning graduate students in a simplified continuous space environment that is relatively free of the distorting effects of networks and other aspects of more differentiated “real-world” environments. This approach can enable instructors to reinforce visually the role of different models and their objective functions. In this simplified setting, students can initially concentrate on the link between these mathematical programming techniques and the spatial nature of the problems being solved. In a short time, students can move into more advanced methods in more differentiated environments. A freeware program entitled NEWLAP was developed to facilitate this approach. This software features a variety of spatial allocation models and their associated constraints that can be applied on both the plane and the sphere. This paper outlines how this software can be used to show alternative solutions using different models on the same data set as well as application of the software in a “real world” problem on a global scale.  相似文献   

13.
Abstract

Research on time and data models for Geographical Information Systems (GIS) has focused mainly in the representation of temporal geographical entities and implementation of temporal databases. Many temporal GIS database structures have been proposed but most of them just provide principles, not the recipe for the design. Owing to the manipulation of the large quantity of geographical information and the slow response time, few implementations exist. This paper presents a relational method of storing and retrieving spatial and temporal topologies. Two-level state topologies are proposed: a state topology for a set of geographical entities and a state topology for a single geographical entity.

From a temporal perspective, these two-level state topologies may also be viewed as two-level time topologies: a time topology for all geographical entities in a GIS database and a time topology for a single geographical entity. Based on these state and time topologies, a detailed storage approach for historical geographical information is provided.  相似文献   

14.
  相似文献   

15.
Urban land use information plays an essential role in a wide variety of urban planning and environmental monitoring processes. During the past few decades, with the rapid technological development of remote sensing (RS), geographic information systems (GIS) and geospatial big data, numerous methods have been developed to identify urban land use at a fine scale. Points-of-interest (POIs) have been widely used to extract information pertaining to urban land use types and functional zones. However, it is difficult to quantify the relationship between spatial distributions of POIs and regional land use types due to a lack of reliable models. Previous methods may ignore abundant spatial features that can be extracted from POIs. In this study, we establish an innovative framework that detects urban land use distributions at the scale of traffic analysis zones (TAZs) by integrating Baidu POIs and a Word2Vec model. This framework was implemented using a Google open-source model of a deep-learning language in 2013. First, data for the Pearl River Delta (PRD) are transformed into a TAZ-POI corpus using a greedy algorithm by considering the spatial distributions of TAZs and inner POIs. Then, high-dimensional characteristic vectors of POIs and TAZs are extracted using the Word2Vec model. Finally, to validate the reliability of the POI/TAZ vectors, we implement a K-Means-based clustering model to analyze correlations between the POI/TAZ vectors and deploy TAZ vectors to identify urban land use types using a random forest algorithm (RFA) model. Compared with some state-of-the-art probabilistic topic models (PTMs), the proposed method can efficiently obtain the highest accuracy (OA = 0.8728, kappa = 0.8399). Moreover, the results can be used to help urban planners to monitor dynamic urban land use and evaluate the impact of urban planning schemes.  相似文献   

16.
Editorial     
Abstract

The analysis of geographical information is compared with other production processes in which a user can only accept an end-product if it meets certain quality requirements. Whereas users are responsible for defining the levels of quality they need to use the results of the analyses of geographical information systems in their work, database managers, experts and modellers could greatly assist users to achieve the quality of results they seek by formalizing information on: (1) data collection, level of resolution and quality; (2) the use of the basic analytical functions of the geographical information system; and (3) the data requirements, sensitivity and error propagation in models. These meta-data could be incorporated in a knowledge base alongside the geographical information system where, together with procedures for on-line error propagation, a user could be advised on the best way to achieve a desired aim. If the analysis showed that the original constellation of data, methods and models could not achieve the aim with the desired quality, the intelligent geographical information system would present a range of alternative strategies—better methods, more data, different data, better models, better model calibration, or better spatial resolution—and their costs by which the user's aims could reasonably be achieved.  相似文献   

17.
Abstract

Integrating local and Indigenous knowledge into land-use planning and the assessment of ecosystems services requires reliable, quantitative data. We tested two approaches to obtain such data by quantifying farmer opinion of different land-covers in Eastern Panama using (1) the analytic hierarchy process (AHP) and (2) a simpler ranking and scoring technique. Both methods produce a set of numerical values reflecting the ability of land-covers to deliver ecological and socio-economic criteria. We present our experience with both methods and offer recommendations for researchers looking to quantify landholder opinion. The AHP survey was relatively long (on average it took 19?min to complete per criterion) and we faced problems with inconsistent responses. In contrast, the ranking and scoring method was much quicker (only 3?min per criterion) and therefore may be more suitable for gathering more data from a larger number of farmers.  相似文献   

18.
Abstract

Representations historically used within GIS assume a world that exists only in the present. Information contained within a spatial database may be added-to or modified over time, but a sense of change or dynamics through time is not maintained. This limitation of current GIS capabilities has recently received substantial attention, given the increasingly urgent need to better understand geographical processes and the cause-and-effect interrelationships between human activities and the environment. Models proposed so-far for the representation of spatiotemporal data are extensions of traditional raster and vector representations that can be seen as location- or feature-based, respectively, and are therefore best organized for performing either location-based or feature-based queries. Neither form is as well-suited for analysing overall temporal relationships of events and patterns of events throughout a geographical area as a temporally-based representation.

In the current paper, a new spatio-temporal data model is proposed that is based on time as its organizational basis, and is thereby intended to facilitate analysis of temporal relationships and patterns of change through time. This model is named the Event-based Spatio Temporal Data Model (ESTDM). It is shown that temporally-based queries relating to locations can be implemented in an efficient and conceptually straightforward manner using ESTDM by describing algorithms for three fundamental temporally-based retrieval tasks based on this model: (1) retrieving location(s) that changed to a given value at a given time, (2) retrieving location(s) that changed to a given value over a given temporal interval, and (3) calculation of the total area that has changed to a given value over a given temporal interval. An empirical comparison of the space efficiency of ESTDM and compressed and uncompressed forms of the ‘snapshot’ model is also given, showing that ESTDM is also a compact representation of spatio-temporal information.  相似文献   

19.
Abstract

Despite the continued importance of regional courses in college curricula and the far-reaching innovations in systematic geography, the formats and organizational structure of regional geography has remained obdurately stable. The persistent demand for a regional course on Africa, together with the disturbing anachronistic curricular materials and our difficulty in finding adequate materials for our class, stimulated us to develop a different approach to teaching African geography. The approach is thematic and is intended to modernize African regional geography, bring it into phase with the rest of the discipline, and emphasize the essential unity of the spatial viewpoint. Five themes are used as the framework for providing an integrating overview of the continent and its problems.  相似文献   

20.
Abstract

Many spatial decision-making problems, such as site selection or land use allocation require the decision-maker to consider the impacts of choice-alternatives along multiple dimensions in order to choose the best alternative. The decision-making process, involving policy priorities, trade-offs, and uncertainties, can be aided by Multiple Criteria Decision making (MCDM) methods. This paper presents a framework for integrating geographical information systems (GIS) and MCDM methods. In this framework the MCDM methods are classified and matched with choice heuristics used by the decision-makers in the presence of competing alternatives and multiple evaluation criteria. Two strategies for integrating GIS with MCDM are proposed. The first strategy suggests linking GIS and MCDM techniques using a file exchange mechanism. The second strategy suggests integrating GIS and MCDM functions using a common database. The paper presents the implementation of the first strategy using PC-ARC/INFO, a file exchange module, and four different MCDM computer programs.  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