Search results

The purpose of this study is to analyse the suitability of photo-sharing platforms, such as Flickr, to extract relevant knowledge on tourists’ spatial movement and point of interest (POI) visitation behaviour and compare the most prominent clustering approaches to identify POIs in various application scenarios.

The study, first, extracts photo metadata from Flickr, such as upload time, location and user. Then, photo uploads are assigned to latent POIs by density-based spatial clustering of applications with noise (DBSCAN) and k-means clustering algorithms. Finally, association rule analysis (FP-growth algorithm) and sequential pattern mining (generalised sequential pattern algorithm) are used to identify tourists’ behavioural patterns.

The approach has been demonstrated for the city of Munich, extracting 13,545 photos for the year 2015. POIs, identified by DBSCAN and k-means clustering, could be meaningfully assigned to well-known POIs. By doing so, both techniques show specific advantages for different usage scenarios. Association rule analysis revealed strong rules (support: 1.0-4.6 per cent; lift: 1.4-32.1 per cent), and sequential pattern mining identified relevant frequent visitation sequences (support: 0.6-1.7 per cent).

As a theoretic contribution, this study comparatively analyses the suitability of different clustering techniques to appropriately identify POIs based on photo upload data as an input to association rule analysis and sequential pattern mining as an alternative but also complementary techniques to analyse tourists’ spatial behaviour.

From a practical perspective, the study highlights that big data sources, such as Flickr, show the potential to effectively substitute traditional data sources for analysing tourists’ spatial behaviour and movement patterns within a destination. Especially, the approach offers the advantage of being fully automatic and executable in a real-time environment.

The study presents an approach to identify POIs by clustering photo uploads on social media platforms and to analyse tourists’ spatial behaviour by association rule analysis and sequential pattern mining. The study gains novel insights into the suitability of different clustering techniques to identify POIs in different application scenarios.

本论文旨在分析图片分享平台Flickr对截取游客空间动线信息和景点（POI）游览行为的适用性, 并且对比最知名的几种聚类分析手段, 以确定不同情况下的POI。

本论文首先从Flickr上摘录下图片大数据, 比如上传时间、地点、用户等。其次, 本论文使用DBSCAN和k-means聚类分析参数来将上传图片分配给POI隐性变量。最后, 本论文采用关联规则挖掘分析（FP-growth参数）和序列样式勘探分析（GSP参数）以确认游客行为模式。

本论文以慕尼黑城市为样本, 截取2015年13,545张图片。POIs由DBSCAN和k-means聚类分析将其分配到有名的POIs。由此, 本论文证明了两种技术对不同用法的各自优势。关联规则挖掘分析显示了显著联系（support：1%−4.6%；lift：1.4%−32.1%）, 序列样式勘探分析确立了相关频率游览次序（support：0.6%−1.7%。

本论文的理论贡献在于, 根据图片数据, 通过对比分析不同聚类分析技术对确立POIs, 并且证明关联规则挖掘分析和序列样式勘探分析各有千秋又互相补充的分析技术以确立游客空间行为。

本论文的现实意义在于, 强调了大数据的来源, 比如Flickr,证明了其对于有效代替传统数据的潜力, 以分析在游客在一个旅游目的地的空间行为和动线模式。特别是这种方法实现了实时自动可操作性等优势。

本论文展示了一种方法, 这种方法通过聚类分析社交媒体上的上传图片以确立POIs, 以及通过关联规则挖掘分析和序列样式勘探分析来分析游客空间行为。本论文对于不同聚类分析以确立不同适用情况下的POIs的确立提出了独到见解。

Accurate prediction of temperature and heat is crucial for the design of various nano/micro devices in engineering. Recently, investigation has been carried out for calculating the heat flux of gas flow using the concept of sliding friction because of the slip velocity at the surface. The purpose of this study is to exetend the concept of sliding friction for various types of nano/micro flows.

A new type of Smoluchowski temperature jump considering the viscous heat generation (sliding friction) has recently been proposed (Le and Vu, 2016b) as an alternative jump condition for the prediction of the surface gas temperature at solid interfaces for high-speed non-equilibrium gas flows. This paper investigated the proposed jump condition for the nano/microflows which has not been done earlier using four cases: 90° bend microchannel pressure-driven flow, nanochannel backward facing step with a pressure-driven flow, nanoscale flat plate and NACA 0012 micro-airfoil. The results are compared with the available direct simulation Monte Carlo results. Also, this paper has demonstrated low-speed preconditioned density-based algorithm for the rarefied gas flows. The algorithm captured even very low Mach numbers of 2.12 × 10⁻⁵.

Based on this study, this paper concludes that the effect of inclusion of sliding friction in improving the thermodynamic prediction is case-dependent. It is shown that its performance depends not only on the slip velocity at the surface but also on the mean free path of the gas molecule and the shear stress at the surface. A pressure jump condition was used along with the new temperature jump condition and it has been found to often improve the prediction of surface flow properties significantly.

This paper extends the concept of using sliding friction at the wall for micro/nano flows. The pressure jump condition was used which has been generally ignored by researchers and has been found to often improve the prediction of surface flow properties. Different flow properties have been studied at the wall apart from only temperature and heat flux, which was not done earlier.

Internet marketing underground industry users refer to people who use technology means to simulate a large number of real consumer behaviors to obtain marketing activities rewards illegally, which leads to increased cost of enterprises and reduced effect of marketing. Therefore, this paper aims to construct a user risk assessment model to identify potential underground industry users to protect the interests of real consumers and reduce the marketing costs of enterprises.

Method feature extraction is based on two aspects. The first aspect is based on traditional statistical characteristics, using density-based spatial clustering of applications with noise clustering method to obtain user-dense regions. According to the total number of users in the region, the corresponding risk level of the receiving address is assigned. So that high-quality address information can be extracted. The second aspect is based on the time period during which users participate in activities, using frequent item set mining to find multiple users with similar operations within the same time period. Extract the behavior flow chart according to the user participation, so that the model can mine the deep relationship between the participating behavior and the underground industry users.

Based on the real underground industry user data set, the features of the data set are extracted by the proposed method. The features are experimentally verified by different models such as random forest, fully-connected layer network, SVM and XGBOST, and the proposed method is comprehensively evaluated. Experimental results show that in the best case, our method can improve the F1-score of traditional models by 55.37%.

This paper investigates the relative importance of static information and dynamic behavior characteristics of users in predicting underground industry users, and whether the absence of features of these categories affects the prediction results. This investigation can go a long way in aiding further research on this subject and found the features which improved the accuracy of predicting underground industry users.

This paper aims to present a methodology for the detection and categorisation of metal powder particles that are partially attached to additively manufactured lattice structures. It proposes a software algorithm to process micro computed tomography (µCT) image data, thereby providing a systematic and formal basis for the design and certification of powder bed fusion lattice structures, as is required for the certification of medical implants.

This paper details the design and development of a software algorithm for the analysis of µCT image data. The algorithm was designed to allow statistical probability of results based on key independent variables. Three data sets with a single unique parameter were input through the algorithm to allow for characterisation and analysis of like data sets.

This paper demonstrates the application of the proposed algorithm with three data sets, presenting a detailed visual rendering derived from the input image data, with the partially attached particles highlighted. Histograms for various geometric attributes are output, and a continuous trend between the three different data sets is highlighted based on the single unique parameter.

This paper presents a novel methodology for non-destructive algorithmic detection and categorisation of partially attached metal powder particles, of which no formal methods exist. This material is available to download as a part of a provided GitHub repository.

This paper aims to highlight how big social data (BSD) and analytics exploitation may help destination management organisations (DMOs) to understand tourist behaviours and destination experiences and images. Gathering data from two different sources, Flickr and Twitter, textual and visual contents are used to perform different analytics tasks to generate insights on tourist behaviour and the affective aspects of the destination image.

This work adopts a method based on a multimodal approach on BSD and analytics, considering multiple BSD sources, different analytics techniques on heterogeneous data types, to obtain complementary results on the Salento region (Italy) case study.

Results show that the generated insights allow DMOs to acquire new knowledge about discovery of unknown clusters of points of interest, identify trends and seasonal patterns of tourist demand, monitor topic and sentiment and identify attractive places. DMOs can exploit insights to address its needs in terms of decision support for the management and development of the destination, the enhancement of destination attractiveness, the shaping of new marketing and communication strategies and the planning of tourist demand within the destination.

The originality of this work is in the use of BSD and analytics techniques for giving DMOs specific insights on a destination in a deep and wide fashion. Collected data are used with a multimodal analytic approach to build tourist characteristics, images, attitudes and preferred destination attributes, which represent for DMOs a unique mean for problem-solving, decision-making, innovation and prediction.

A wide number of technologies are currently in store to harness the challenges posed by pandemic situations. As such diseases transmit by way of person-to-person contact or by any other means, the World Health Organization had recommended location tracking and tracing of people either infected or contacted with the patients as one of the standard operating procedures and has also outlined protocols for incident management. Government agencies use different inputs such as smartphone signals and details from the respondent to prepare the travel log of patients. Each and every event of their trace such as stay points, revisit locations and meeting points is important. More trained staffs and tools are required under the traditional system of contact tracing. At the time of the spiralling patient count, the time-bound tracing of primary and secondary contacts may not be possible, and there are chances of human errors as well. In this context, the purpose of this paper is to propose an algorithm called SemTraClus-Tracer, an efficient approach for computing the movement of individuals and analysing the possibility of pandemic spread and vulnerability of the locations.

Pandemic situations push the world into existential crises. In this context, this paper proposes an algorithm called SemTraClus-Tracer, an efficient approach for computing the movement of individuals and analysing the possibility of pandemic spread and vulnerability of the locations. By exploring the daily mobility and activities of the general public, the system identifies multiple levels of contacts with respect to an infected person and extracts semantic information by considering vital factors that can induce virus spread. It grades different geographic locations according to a measure called weightage of participation so that vulnerable locations can be easily identified. This paper gives directions on the advantages of using spatio-temporal aggregate queries for extracting general characteristics of social mobility. The system also facilitates room for the generation of various information by combing through the medical reports of the patients.

It is identified that context of movement is important; hence, the existing SemTraClus algorithm is modified by accounting for four important factors such as stay point, contact presence, stay time of primary contacts and waypoint severity. The priority level can be reconfigured according to the interest of authority. This approach reduces the overwhelming task of contact tracing. Different functionalities provided by the system are also explained. As the real data set is not available, experiments are conducted with similar data and results are shown for different types of journeys in different geographical locations. The proposed method efficiently handles computational movement and activity analysis by incorporating various relevant semantics of trajectories. The incorporation of cluster-based aggregate queries in the model do away with the computational headache of processing the entire mobility data.

As the trajectory of patients is not available, the authors have used the standard data sets for experimentation, which serve the purpose.

This paper proposes a framework infrastructure that allows the emergency response team to grab multiple information based on the tracked mobility details of a patient and facilitates room for various activities for the mitigation of pandemics such as the prediction of hotspots, identification of stay locations and suggestion of possible locations of primary and secondary contacts, creation of clusters of hotspots and identification of nearby medical assistance. The system provides an efficient way of activity analysis by computing the mobility of people and identifying features of geographical locations where people travelled. While formulating the framework, the authors have reviewed many different implementation plans and protocols and arrived at the conclusion that the core strategy followed is more or less the same. For the sake of a reference model, the Indian scenario is adopted for defining the concepts.

The purpose of this paper is to solve unsteady compressible Navier–Stokes equations without the commonly used dual-time loop. The authors would like to use an adaptive time-stepping (ATS)-based local error control instead of CFL-based time-stepping technique. Also, an all-speed flow algorithm is implemented with simple low dissipation AUSM convective scheme, which can be computed without preconditioning which in general destroys the time accuracy.

In transient flow computations, the time-step is generally determined from the CFL condition. In this paper, the authors demonstrate the usefulness of ATS based on local time-stepping previously used extensively in ordinary differential equations (ODE) integration. This method is implemented in an implicit framework to ensure the numerical domain of dependence always contains the physical domain of dependence.

In this paper, the authors limit their focus to capture the unsteady physics for three cases: Sod’s shock-tube problem, Stokes’ second problem and a circular cylinder. The use of ATS with local truncation error control enables the solver to use the maximum allowable time-step, for the prescribed tolerance of error. The algorithm is also capable of converging very rapidly to the steady state (if there is any) after the initial transient phase. The authors present here only the first-order time-stepping scheme. An algorithmic comparison is made between the proposed adaptive time-stepping method and the commonly used dual time-stepping approach that indicates the former will be more efficient.

The original method of ATS based on local error control is used extensively in ODE integration, whereas, this method is not so popular in the computational fluid dynamics (CFD) community. In this paper, the authors investigate its use in the unsteady CFD computations. The authors hope that it would provide CFD researchers with an algorithm based on an adaptive time-stepping approach for unsteady calculations.

Service-oriented architecture is an emerging software architecture, in which web service (WS) plays a crucial role. In this architecture, the task of WS composition and verification is required when handling complex requirement of services from users. When the number of WS becomes very huge in practice, the complexity of the composition and verification is also correspondingly high. In this paper, the authors aim to propose a logic-based clustering approach to solve this problem by separating the original repository of WS into clusters. Moreover, they also propose a so-called quality-controlled clustering approach to ensure the quality of generated clusters in a reasonable execution time.

The approach represents WSs as logical formulas on which the authors conduct the clustering task. They also combine two most popular clustering approaches of hierarchical agglomerative clustering (HAC) and k-means to ensure the quality of generated clusters.

This logic-based clustering approach really helps to increase the performance of the WS composition and verification significantly. Furthermore, the logic-based approach helps us to maintain the soundness and completeness of the composition solution. Eventually, the quality-controlled strategy can ensure the quality of generated clusters in low complexity time.

The work discussed in this paper is just implemented as a research tool known as WSCOVER. More work is needed to make it a practical and usable system for real life applications.

In this paper, the authors propose a logic-based paradigm to represent and cluster WSs. Moreover, they also propose an approach of quality-controlled clustering which combines and takes advantages of two most popular clustering approaches of HAC and k-means.

The K-means (KM) clustering algorithm is extremely responsive to the selection of initial centroids since the initial centroid of clusters determines computational effectiveness, efficiency and local optima issues. Numerous initialization strategies are to overcome these problems through the random and deterministic selection of initial centroids. The random initialization strategy suffers from local optimization issues with the worst clustering performance, while the deterministic initialization strategy achieves high computational cost. Big data clustering aims to reduce computation costs and improve cluster efficiency. The objective of this study is to achieve a better initial centroid for big data clustering on business management data without using random and deterministic initialization that avoids local optima and improves clustering efficiency with effectiveness in terms of cluster quality, computation cost, data comparisons and iterations on a single machine.

This study presents the Normal Distribution Probability Density (NDPD) algorithm for big data clustering on a single machine to solve business management-related clustering issues. The NDPDKM algorithm resolves the KM clustering problem by probability density of each data point. The NDPDKM algorithm first identifies the most probable density data points by using the mean and standard deviation of the datasets through normal probability density. Thereafter, the NDPDKM determines K initial centroid by using sorting and linear systematic sampling heuristics.

The performance of the proposed algorithm is compared with KM, KM++, Var-Part, Murat-KM, Mean-KM and Sort-KM algorithms through Davies Bouldin score, Silhouette coefficient, SD Validity, S_Dbw Validity, Number of Iterations and CPU time validation indices on eight real business datasets. The experimental evaluation demonstrates that the NDPDKM algorithm reduces iterations, local optima, computing costs, and improves cluster performance, effectiveness, efficiency with stable convergence as compared to other algorithms. The NDPDKM algorithm minimizes the average computing time up to 34.83%, 90.28%, 71.83%, 92.67%, 69.53% and 76.03%, and reduces the average iterations up to 40.32%, 44.06%, 32.02%, 62.78%, 19.07% and 36.74% with reference to KM, KM++, Var-Part, Murat-KM, Mean-KM and Sort-KM algorithms.

The KM algorithm is the most widely used partitional clustering approach in data mining techniques that extract hidden knowledge, patterns and trends for decision-making strategies in business data. Business analytics is one of the applications of big data clustering where KM clustering is useful for the various subcategories of business analytics such as customer segmentation analysis, employee salary and performance analysis, document searching, delivery optimization, discount and offer analysis, chaplain management, manufacturing analysis, productivity analysis, specialized employee and investor searching and other decision-making strategies in business.

The electronic nose is an array of chemical or gas sensors and associated with a pattern-recognition framework competent in identifying and classifying odorant or non-odorant and simple or complex gases. Despite more than 30 years of research, the robust e-nose device is still limited. Most of the challenges towards reliable e-nose devices are associated with the non-stationary environment and non-stationary sensor behaviour. Data distribution of sensor array response evolves with time, referred to as non-stationarity. The purpose of this paper is to provide a comprehensive introduction to challenges related to non-stationarity in e-nose design and to review the existing literature from an application, system and algorithm perspective to provide an integrated and practical view.

The authors discuss the non-stationary data in general and the challenges related to the non-stationarity environment in e-nose design or non-stationary sensor behaviour. The challenges are categorised and discussed with the perspective of learning with data obtained from the sensor systems. Later, the e-nose technology is reviewed with the system, application and algorithmic point of view to discuss the current status.

The discussed challenges in e-nose design will be beneficial for researchers, as well as practitioners as it presents a comprehensive view on multiple aspects of non-stationary learning, system, algorithms and applications for e-nose. The paper presents a review of the pattern-recognition techniques, public data sets that are commonly referred to as olfactory research. Generic techniques for learning in the non-stationary environment are also presented. The authors discuss the future direction of research and major open problems related to handling non-stationarity in e-nose design.

The authors first time review the existing literature related to learning with e-nose in a non-stationary environment and existing generic pattern-recognition algorithms for learning in the non-stationary environment to bridge the gap between these two. The authors also present details of publicly available sensor array data sets, which will benefit the upcoming researchers in this field. The authors further emphasise several open problems and future directions, which should be considered to provide efficient solutions that can handle non-stationarity to make e-nose the next everyday device.