AI-driven landscape values mapping.

Abstract. Individual travel trajectories denote a series of places people visit along the time. These places (e.g., home, workspace, and park) reflect people’s corresponding activities (e.g., dwelling, work, and entertainment), which are discussed as semantic knowledge and could be implicit under raw data (Yan et al. 2013, Cai et al. 2016). Traditional survey data directly describe people’ activities at certain places, while costing tremendous labors and resources (Huang and Wong 2016). GPS data such as taxi logs record exact origin-destination pairs as well as people’s stay time along the way, from which semantics can be easily inferred combining with geographical context data (Yan et al. 2013). Research has been done to understand the activity sequences indicated by either individual or collective spatiotemporal (ST) travel trajectories using those dense data. Different models are proposed for trajectory mining and activity inference, including location categorization, frequent region detection, and so on (Njoo et al. 2015). A typical method for matching a location or region with a known activity type is to detect stay points and stay intervals of trajectories and to find geographical context of these stay occurrences (Furtado et al. 2013, Njoo et al. 2015, Beber et al. 2016, Beber et al. 2017).However, limited progress has been made to mine semantics of trajectory data collected from social media platforms. Specifically, detection of stay points and their intervals could be inaccurate using online trajectories because of data sparsity. Huang et al. (2014) define the notion of activity zone to detect activity types from digital footprints. In this method, individual travel trajectories first are aggregated using spatial clustering method such as density-based spatial clustering of applications with noise (DBSCAN). Then produced clusters are classified based on a regional land use map and Google Places application programming interface (API). Such land use data are only published at specific places, such as the state cartography office’s website at University of Wisconsin-Madison. Researchers need to search for those data based on their study area. Moreover, while major land use maps can be searched for large areas such as the whole United States, detailed land use data for statewide or citywide areas are made in diverse standards, which adds extra work to classify activity zones consistently. Besides, Google Places API is a service that Google opened for developers and will return information about a place, given the place location (e.g., address or GPS coordinates), in the search request. However, API keys need to be generated before we can use these interfaces and each user can only make a limited number of free-charged requests every day (i.e., 1,000 requests per 24 hours period). In sum, previous methods to detect activity zone types using social media data are not sufficient and can hardly achieve effective data fusion. Comparing to the high cost of using officially published dataset, emerging Volunteered Geographic Information (VGI) data offer an alternative to infer the types of an individual’s activities performed in each zone (i.e., cluster).Using geo-tagged tweets as an example, this research proposes a framework for mining social media data, detecting individual semantic travel trajectories, and individual representative daily travel trajectory paths by fusing with VGI data, specifically OpenStreetMap (OSM) datasets. First, inactive users and abnormal users (e.g., users representing a company with account being shared by many employees) are removed through data pre-processing (Step 1 in Figure 1). Next, a multi-scale spatial clustering method is developed to aggregate online trajectories captured through geo-tagged tweets of a group of users into collective spatial hot-spots (i.e., activity zones; Step 2). By integrating multiple OSM datasets the activity type (e.g., dwelling, service, transportation and work) of each collective zone then can be identified (Step 3). Each geo-tagged tweet of an individual, represented as a ST point, is then attached with a collective activity zone that either includes or overlaps a buffer zone of the ST point. Herein, the buffer zone is generated by using the point as the centroid and a predefined threshold as the radius. Given an individual’s ST points with semantics (i.e., activity type information) derived from the attached collective activity zone, a semantic activity clustering method is then developed to detect daily representative activity clusters of the individual (Step 4). Finally, individual representative daily semantic travel trajectory paths (i.e., semantic travel trajectory, defined as chronological travel activity sequences) are constructed between every two subsequent activity clusters (Step 5). Experiments with the historic geo-tagged tweets collected within Madison, Wisconsin reveal that: 1) The proposed method can detect most significant activity zones with accurate zone types identified (Figure 2); and 2) The semantic activity clustering method based on the derived activity zones can aggregate individual travel trajectories into activity clusters more efficiently comparing to DBSCAN and varying DBSCAN (VDBSCAN).

Investing in the stock market is challenged by high volatility, which often leads to inaccurate price predictions. Prediction models often struggle to handle the fluctuation phenomenon and produce unstable forecasts. This study aims to predict stock prices in three banks, namely PT Bank Central Asia Tbk (BBCA), PT Bank Rakyat Indonesia (Persero) Tbk (BBRI), and PT Bank Mandiri (Persero) Tbk (BMRI) using Long Short-Term Memory (LSTM) with the integration of Density-Based Spatial Clustering of Applications with Noise (DBSCAN) for anomaly detection. DBSCAN is applied with an epsilon (ε) of 0.5 and a minimum of 5 samples using Euclidean distance. The LSTM model consists of two hidden layers with 50 units, optimized using Adam, and applying the Mean Squared Error (MSE) loss function. The results show that DBSCAN improves prediction accuracy under several conditions. For BBCA stock, the lowest MSE was 0.003 at the 2nd fold with DBSCAN compared to 0.006 without DBSCAN. For BMRI stock achieved an MSE of 0.003 at the 4th fold with DBSCAN, while the 5th fold without DBSCAN obtained 0.000. For BBRI stock showed the best MSE of 0.003 at the 2nd fold with DBSCAN and the 5th fold without DBSCAN. These results show that the integration of DBSCAN can improve prediction especially when extreme price fluctuations occur. This research contributes to the development of stock price prediction methods that can be one of the benchmarks for investors before making decisions so that they do not experience losses.

A significant challenge for the growing world of data is to analyze, classify and manipulate spatial data. The challenge starts with the clustering process, which can be defined to characterize the spatial data with their relative properties in different groups or classes. This process can be performed using many different methods like grids, density, hierarchical and others. Among all these methods, the use of density for grouping leads to a lower noise data in result, which is called Density-Based Spatial Clustering of Applications with Noise (DBSCAN). The DBSCAN algorithm defines a data set in a group and separates the group from the other groups based on the density of the data surrounding the selection of data points. These data points and the density of the data are calculated depending on two parameters. One parameter is used as the radius of the data point to find the neighborhood data points. Another parameter is used to identify the noise in the collected data by keeping the minimum number of data points for the data density. Like other popular method k-means, DBSCAN does not require any input of the cluster number. It can sort the data set with the number of clusters according to data density. The purpose of this article is to explain the Efficient Density-based Spatial Clustering of Applications with Noise (DBSCAN) using a sample of data set, compare the results, identify the constraints, and suggest some possible solutions.

In the context of the mid-late development of China’s urbanization, promoting sustainable urban development and giving full play to urban potential have become a social focus, which is of enormous practical significance for the study of urban spatial pattern. Based on such Internet data as a map’s Point of Interest (POI), this paper studies the spatial distribution pattern and clustering characteristics of POIs of four categories of service facilities in Chengdu of Sichuan Province, including catering, shopping, transportation, scientific, educational, and cultural services, by means of spatial data mining technologies such as dimensional autocorrelation analysis and DBSCAN clustering. Global spatial autocorrelation is used to study the correlation between an index of a certain element and itself (univariate) or another index of an adjacent element (bivariate); partial spatial autocorrelation is used to identify characteristics of spatial clustering or spatial anomaly distribution of geographical elements. DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is able to detect clusters of any shape without prior knowledge. The final step is to carry out quantitative analysis and reveal the distribution characteristics and coupling effects of spatial patterns. According to the results, (1) the spatial distribution of POIs of all service facilities is significantly polarized, as they are concentrated in the old city, and the trend of suburbanization is indistinctive, showing three characteristics, namely, central driving, traffic accessibility, and dependence on population activity; (2) the spatial distribution of POIs of the four categories of service facilities is featured by the pattern of “one center, multiple clusters,” where “one center” mainly covers the area within the first ring road and partial region between the first ring road and the third ring road, while “multiple clusters” are mainly distributed in the well-developed areas in the second circle of Chengdu, such as Wenjiang District and Shuangliu District; and (3) there is a significant correlation between any two categories of POIs. Highly mixed multifunctional areas are mainly distributed in the urban center, while service industry is less aggregated in urban fringe areas, and most of them are single-functional or dual-functional regions.

In modern opportunistic networks, network operations can be improved through knowledge of spatial information of low and high density areas, predictions of the mobility of nodes in the space, as well as the spatial distribution of nodes. Such information can be used to adapt forwarding decisions. In this paper, we introduce an efficient opportunistic spatial clustering algorithm, OPSCAN (Opportunistic Spatial Clustering of Applications with Noise). Based on DBSCAN (Density-Based Spatial Clustering of Applications with Noise), a density-based clustering algorithm that discovers arbitrary-shaped clusters in a dataset and distinguishes noise points. OPSCAN is capable of clustering network nodes into high density clusters, while maintaining sparse areas of nodes between clusters. Clusters share spatial information of the network such as area density, mobility statistics and information about other clusters and their nodes. Knowledge of edge nodes in the clusters is also made available for utilization in more efficient forwarding decisions. Simulations show that our algorithm is capable of producing dense, homogeneous clusters and accurately outlining cluster edges. We have used the Silhouette Coefficient to measure cluster homogeneity against density-based clustering algorithms DBSCAN and ST-DBSCAN (Spatial-Temporal DBSCAN), a DBSCAN-based spatial-temporal variant on GeoLife dataset. We have found OPSCAN outperforms DBSCAN by a coefficient of 0.81 to 0.73 for the same minimum distance, under-performing ST-DBSCAN by 0.87 to 0.81 for that distance. OPSCAN requires only two inputs as compared to four for ST-DBSCAN. As the distance parameter is increased, OPSCAN produces homogeneous clusters more closely to ST-DBSCAN.

The density-based spatial clustering of applications with noise (DBSCAN) is able to cluster arbitrarily structured datasets. However, the clustering result of this algorithm is exceptionally sensitive to the neighborhood radius (Eps) and noise points, and it is hard to obtain the best result quickly and accurately with it. To address this issue, a parameter-adaptive DBSCAN clustering algorithm based on the Sparrow Search Algorithm (SSA), referred to as SSA-DBSCAN, is proposed. This method leverages the local fast search ability of SSA, using the optimal number of clusters and the silhouette coefficient of the dataset as the objective functions to iteratively optimize and select the two input parameters of DBSCAN. This avoids the adverse impact of manually inputting parameters, enabling adaptive clustering with DBSCAN. Experiments on typical synthetic datasets, UCI (University of California, Irvine) real-world datasets, and image segmentation tasks have validated the effectiveness of the SSA-DBSCAN algorithm. Comparative analysis with DBSCAN and other related optimization algorithms demonstrates the clustering performance of SSA-DBSCAN.

Most clustering algorithms considering spatial characteristics of data have been developed based on the geological location of observations. Density-based spatial clustering of applications with noise (DBSCAN) provides arbitrarily shaped clusters grouping a set of observations which are closely packed together and noise detecting outliers which lie alone in low-density regions. A distance measure for DBSCAN is Euclidean distance, which is the standard measure of distance and especially suitable to handle continuous variables. To handle both categorical and continuous variables simultaneously, other measures are required to compute distance for various types of variables. Thus, we propose DBSCAN algorithm using Gower distance. We provide numerical results on spatial and non-spatial setup comparing DBSCAN methods with Euclidean and Gower distance and we apply this method to land price data and migraine treatments data. DBSCAN using Gower distance has a reasonable method and gives comparably stable results.

In the urban transportation system, the unbalanced relationship between taxi demand and the number of running taxis reduces the drivers' income and the levels of passengers' satisfaction. With the help of vehicular global positioning system (GPS) data, the taxi demand distribution of city can be analyzed to provide advice for drivers. A clustering algorithm called Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is suitable for discovering demand hotspots. However, the execution efficiency is still a big challenge when DBSCAN is applied on big databases. In this paper, we propose an improved density-based clustering algorithm called Grid and Kd-tree for DBSCAN (GD-DBSCAN), which integrates partitioning method with kd-tree structure to improve the computational performance of DBSCAN. Furthermore, this algorithm can take advantages of multi cores and shared memory to parallelize related functions. The experiment shows GD- DBSCAN is efficient, it has an improvement of at least 10% in performance compared with DBSCAN.

In recent years, machine learning models have exhibited excellent performance and far-reaching impact across domains such as fraud detection in finance, recommendation systems in e-commerce, medical imaging in healthcare, agricultural forecasting, social engagement, image classification, sentiment analysis in social media network analysis. This research explores how advanced machine learning techniques, leveraging social media data for image classification, can be used to gain deeper insights into public engagement with urban wild spaces. The study follows a two-step methodology: first, scraping image data from Instagram, Facebook, and Flickr using hashtag-based techniques focused on urban wild spaces; second, developing an experimental pipeline using Convolutional Neural Networks (CNN), Density-Based Spatial Clustering of Applications with Noise (DBSCAN), and Convolutional Autoencoders (CAE) to classify and evaluate the scrapped social media data. Evaluation was based on precision, recall, F-measure, and accuracy metrics. Across all three platforms, CAE consistently outperformed CNN and DBSCAN, achieving peak accuracies of 74.8% on Flickr, 70.4% on Instagram, and 62.9% on Facebook, along with balanced F-measures and high recall. CNN showed the highest precision, reaching 98.4% on Flickr, while DBSCAN provided moderate results. These findings show that machine learning effectively filters noisy data and reveals how people engage with urban wild spaces, offering valuable insights for urban planning and ecology.

In the past researches, industrial cluster mainly focused on single or particular industry and less on spatial industrial structure and mutual relations. Industrial cluster could generate three kinds of spillover effects, including knowledge, labor market pooling, and input sharing. In addition, industrial cluster indeed benefits industry development. To fully control the status and characteristics of district industrial cluster can facilitate to improve the competitive ascendancy of district industry. The related researches on industrial spatial cluster were of great significance for setting up industrial policies and promoting district economic development. In this study, an improved model, GeoSOM, that combines DBSCAN (Density-Based Spatial Clustering of Applications with Noise) and SOM (Self-Organizing Map) was developed for analyzing industrial cluster. Different from former distance-based algorithm for industrial cluster, the proposed GeoSOM model can calculate spatial characteristics between firms based on DBSCAN algorithm and evaluate the similarity between firms based on SOM clustering analysis. The demonstrative data sets, the manufacturers around Taichung County in Taiwan, were analyzed for verifying the practicability of the proposed model. The analyzed results indicate that GeoSOM is suitable for evaluating spatial industrial cluster.

Most human behaviors are related to hot regions. The regularity of region transition is always behind the location transitions. DBSCAN (density-based spatial clustering of applications with noise) is a kind of density-based clustering method which is suitable for spatial clustering. RNN (recurrent neural network) is a kind of network which has a excellent capacity of capturing the sequential transitions. In this paper, we propose combining DBSCAN with the RNN-based model DeepMove to predict human mobility. DBSCAN is applied to the corresponding coordinates of all non-repeating discrete locations to obtain the region identification that represents the hot region or non-hot region of the users for the specific dataset. Having inserted the region identification into each record, the data is fed into DeepMove for training. An experiment is conducted on a real-life dataset Foursquare, of which the result shows it improves top-1 accuracy by 12.9% compared to single DeepMove.

Spatial analysis is an important means of mining floating car trajectory information, and clustering method and density analysis are common methods among them. The choice of the clustering method affects the accuracy and time efficiency of the analysis results. Therefore, clarifying the principles and characteristics of each method is the primary prerequisite for problem solving. Taking four representative spatial analysis methods—KMeans, Density-Based Spatial Clustering of Applications with Noise (DBSCAN), Clustering by Fast Search and Find of Density Peaks (CFSFDP), and Kernel Density Estimation (KDE)—as examples, combined with the hotspot spatiotemporal mining problem of taxi trajectory, through quantitative analysis and experimental verification, it is found that DBSCAN and KDE algorithms have strong hotspot discovery capabilities, but the heat regions’ shape of DBSCAN is found to be relatively more robust. DBSCAN and CFSFDP can achieve high spatial accuracy in calculating the entrance and exit position of a Point of Interest (POI). KDE and DBSCAN are more suitable for the classification of heat index. When the dataset scale is similar, KMeans has the highest operating efficiency, while CFSFDP and KDE are inferior. This paper resolves to a certain extent the lack of scientific basis for selecting spatial analysis methods in current research. The conclusions drawn in this paper can provide technical support and act as a reference for the selection of methods to solve the taxi trajectory mining problem.

Anomaly detection concerns identifying anomalous observations or patterns that are a deviation from the dataset's expected behaviour. The detection of anomalies has significant and practical applications in several industrial domains such as public health, finance, Information Technology (IT), security, medical, energy, and climate studies. Density-Based Spatial Clustering of Applications with Noise (DBSCAN) Algorithm is a density-based clustering algorithm with the capability of identifying anomalous data. In this paper, a modified DBSCAN algorithm is proposed for anomaly detection in time-series data with seasonality. For experimental evaluation, a monthly temperature dataset was employed and the analysis set forth the advantages of the modified DBSCAN over the standard DBSCAN algorithm for the seasonal datasets. From the result analysis, we may conclude that DBSCAN is used for finding the anomalies in a dataset but fails to find local anomalies in seasonal data. The proposed Modified DBSCAN approach helps to find both the global and local anomalies from the seasonal data. Using normal DBSCAN we are able to get 19 (2.16%) anomaly points. While using the modified approach for DBSCAN we are able to get 42 (4.79%) anomaly points. In comparison we can say that we are able to get 2.11% more anomalies using the modified DBSCAN approach. Hence, the proposed Modified DBSCAN algorithm outperforms in comparison with the DBSCAN algorithm to find local anomalies.

DBSCAN (density-based spatial clustering of applications with noise) is one of the most widely used density-based clustering algorithms, which can find arbitrary shapes of clusters, determine the number of clusters, and identify noise samples automatically. However, the performance of DBSCAN is significantly limited as it is quite sensitive to the parameters of eps and MinPts. Eps represents the eps-neighborhood and MinPts stands for a minimum number of points. Additionally, a dataset with large variations in densities will probably trap the DBSCAN because its parameters are fixed. In order to overcome these limitations, we propose a new density-clustering algorithm called GNN-DBSCAN which uses an adaptive Grid to divide the dataset and defines local core samples by using the Nearest Neighbor. With the help of grid, the dataset space will be divided into a finite number of cells. After that, the nearest neighbor lying in every filled cell and adjacent filled cells are defined as the local core samples. Then, GNN-DBSCAN obtains global core samples by enhancing and screening local core samples. In this way, our algorithm can identify higher-quality core samples than DBSCAN. Lastly, give these global core samples and use dynamic radius based on k-nearest neighbors to cluster the datasets. Dynamic radius can overcome the problems of DBSCAN caused by its fixed parameter eps. Therefore, our method can perform better on dataset with large variations in densities. Experiments on synthetic and real-world datasets were conducted. The results indicate that the average Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Adjusted Mutual Information (AMI) and V-measure of our proposed algorithm outperform the existing algorithm DBSCAN, DPC, ADBSCAN, and HDBSCAN.

Clustering is a task that aims to grouping data objects into several groups. DBSCAN is a density-based clustering method. However, it requires two parameters and these two parameters are hard to decide. Also, DBSCAN has difficulties in finding clusters when the density changes in the dataset. In this paper, we modify the original DBSCAN to make it able to determine the appropriate eps values according to data distribution and to cluster when the density varies among dataset. The main idea is to run DBSCAN with different eps and Minpts values. We also modified the calculation of the Minpts so that DBSCAN can have better clustering results. We did several experiments to evaluate the performance. The results suggest that our proposed DBSCAN can automatically decide the appropriate eps and Minpts values and can detect clusters with different density-levels.