Search results

Class-based storage has been studied extensively and proved to be an efficient storage policy. However, few literature addressed how to cluster stuck items for class-based storage. The purpose of this paper is to develop a constrained clustering method integrated with principal component analysis (PCA) to meet the need of clustering stored items with the consideration of practical storage constraints.

In order to consider item characteristic and the associated storage restrictions, the must-link and cannot-link constraints were constructed to meet the storage requirement. The cube-per-order index (COI) which has been used for location assignment in class-based warehouse was analyzed by PCA. The proposed constrained clustering method utilizes the principal component loadings as item sub-group features to identify COI distribution of item sub-groups. The clustering results are then used for allocating storage by using the heuristic assignment model based on COI.

The clustering result showed that the proposed method was able to provide better compactness among item clusters. The simulated result also shows the new location assignment by the proposed method was able to improve the retrieval efficiency by 33 percent.

While number of items in warehouse is tremendously large, the human intervention on revealing storage constraints is going to be impossible. The developed method can be easily fit in to solve the problem no matter what the size of the data is.

The case study demonstrated an example of practical location assignment problem with constraints. This paper also sheds a light on developing a data clustering method which can be directly applied on solving the practical data analysis issues.

The aim of this study is to propose a deep neural network (DNN) method that uses side information to improve clustering results for big datasets; also, the authors show that applying this information improves the performance of clustering and also increase the speed of the network training convergence.

In data mining, semisupervised learning is an interesting approach because good performance can be achieved with a small subset of labeled data; one reason is that the data labeling is expensive, and semisupervised learning does not need all labels. One type of semisupervised learning is constrained clustering; this type of learning does not use class labels for clustering. Instead, it uses information of some pairs of instances (side information), and these instances maybe are in the same cluster (must-link [ML]) or in different clusters (cannot-link [CL]). Constrained clustering was studied extensively; however, little works have focused on constrained clustering for big datasets. In this paper, the authors have presented a constrained clustering for big datasets, and the method uses a DNN. The authors inject the constraints (ML and CL) to this DNN to promote the clustering performance and call it constrained deep embedded clustering (CDEC). In this manner, an autoencoder was implemented to elicit informative low dimensional features in the latent space and then retrain the encoder network using a proposed Kullback–Leibler divergence objective function, which captures the constraints in order to cluster the projected samples. The proposed CDEC has been compared with the adversarial autoencoder, constrained 1-spectral clustering and autoencoder + k-means was applied to the known MNIST, Reuters-10k and USPS datasets, and their performance were assessed in terms of clustering accuracy. Empirical results confirmed the statistical superiority of CDEC in terms of clustering accuracy to the counterparts.

First of all, this is the first DNN-constrained clustering that uses side information to improve the performance of clustering without using labels in big datasets with high dimension. Second, the author defined a formula to inject side information to the DNN. Third, the proposed method improves clustering performance and network convergence speed.

Little works have focused on constrained clustering for big datasets; also, the studies in DNNs for clustering, with specific loss function that simultaneously extract features and clustering the data, are rare. The method improves the performance of big data clustering without using labels, and it is important because the data labeling is expensive and time-consuming, especially for big datasets.

Examines the processes of cluster analysis and describes them using an example of benefit segmentation, and also discusses other applications suggesting new directions of research in related fields. Bases an example study with 200 early respondents to a survey into sixth formers' choice of degree course, in which students were given 23 criteria which related to their course choice. Comparisons of likeness using Euclidean distance measures were employed. Uses also importance ratings given by three drivers to characteristics of new cars. Proposes that hierarchical clustering can be criticised when used to cluster data that is not naturally hierarchical, but other procedures have similar failings. Posits that clumping and optimisation in conjunction with hierarchical clustering offer the greater potential. Concludes that cluster analysis is a flexible tool, which provides a number of opportunities for marketing, and it is an appealing and simple idea ‐ but there are many technical questions that a researcher must ask before it is used.

Due to the increasing size and complexity of construction projects, construction engineering and management involves the coordination of many complex and dynamic processes and relies on the analysis of uncertain, imprecise and incomplete information, including subjective and linguistically expressed information. Various modelling and computing techniques have been used by construction researchers and applied to practical construction problems in order to overcome these challenges, including fuzzy hybrid techniques. Fuzzy hybrid techniques combine the human-like reasoning capabilities of fuzzy logic with the capabilities of other techniques, such as optimization, machine learning, multi-criteria decision-making (MCDM) and simulation, to capitalise on their strengths and overcome their limitations. Based on a review of construction literature, this chapter identifies the most common types of fuzzy hybrid techniques applied to construction problems and reviews selected papers in each category of fuzzy hybrid technique to illustrate their capabilities for addressing construction challenges. Finally, this chapter discusses areas for future development of fuzzy hybrid techniques that will increase their capabilities for solving construction-related problems. The contributions of this chapter are threefold: (1) the limitations of some standard techniques for solving construction problems are discussed, as are the ways that fuzzy methods have been hybridized with these techniques in order to address their limitations; (2) a review of existing applications of fuzzy hybrid techniques in construction is provided in order to illustrate the capabilities of these techniques for solving a variety of construction problems and (3) potential improvements in each category of fuzzy hybrid technique in construction are provided, as areas for future research.

This paper aims to construct rental housing indices and identify market segmentation for more effective property-management strategies.

The hedonic model was employed to construct the rental indices. Using the k-means++ and REDCAP (Regionalisation with Dynamically Constrained Agglomerative Clustering and Partitioning) approaches, the authors conducted clustering analysis and identified different market segmentation. The empirical study relied on the database of 80,212 actual rental transactions in Beijing, China, spanning 2016–2018.

Rental housing market segmentation may distribute across administrative boundaries. Properly segmented indices could provide a better account for the heterogeneity and spatial continuity of rental housing and as well be crucial for effective property management.

Residential rent might not only vary over space but also interplays with housing price. It would be worth studying how the rental market functions together with the owner-occupied sector in the future.

Residential rental indices are of great importance for policymakers to be able to evaluate housing policies and for property managers to implement competitive strategies in the rental market. Their constructions largely depend on the analysis of market segmentation, a trade-off between housing spatial heterogeneity and continuity.

This paper fills the gap in knowledge concerning segmented rental indices construction, particularly in China. The spatial constrained clustering approach (REDCAP) was also initially introduced to identify regionalised market segmentation due to its superior performance.

It is crucial for information retrieval systems to learn more about what users search for in order to fulfil the intent of searches. This paper introduces query taxonomy generation, which attempts to organise users’ queries into a hierarchical structure of topic classes. Such a query taxonomy provides a basis for the in‐depth analysis of users’ queries on a larger scale and can benefit many information retrieval systems. The proposed approach to this problem consists of two computational processes: hierarchical query clustering to generate a query taxonomy from scratch, and query categorisation to place newly‐arrived queries into the taxonomy. The results of the preliminary experiment have shown the potential of the proposed approach in generating taxonomies for queries, which may be useful in various Web information retrieval applications.

The purpose of this paper is to provide a review of the issues related to cluster analysis, one of the most important and primitive activities of human beings, and of the advances made in recent years.

The paper investigates the clustering algorithms rooted in machine learning, computer science, statistics, and computational intelligence.

The paper reviews the basic issues of cluster analysis and discusses the recent advances of clustering algorithms in scalability, robustness, visualization, irregular cluster shape detection, and so on.

The paper presents a comprehensive and systematic survey of cluster analysis and emphasizes its recent efforts in order to meet the challenges caused by the glut of complicated data from a wide variety of communities.

In particular, this paper aims to systematically analyze a few prominent wireless sensor network (WSN) clustering routing protocols and compare these different approaches according to the taxonomy and several significant metrics.

In this paper, the authors have summarized recent research results on data routing in sensor networks and classified the approaches into four main categories, namely, data-centric, hierarchical, location-based and quality of service (QoS)-aware, and the authors have discussed the effect of node placement strategies on the operation and performance of WSNs.

Performance-controlled planned networks, where placement and routing must be intertwined and everything from delays to throughput to energy requirements is well-defined and relevant, is an interesting subject of current and future research. Real-time, deadline guarantees and their relationship with routing, mac-layer, duty-cycles and other protocol stack issues are interesting issues that would benefit from further research.

This paper aims to examine how the regional variable in country conflict modeling affects forecast accuracy and identifies a methodology to further improve the predictions.

This paper uses statistical learning methods to both evaluate the quantity of data for clustering countries along with quantifying accuracy according to the number of clusters used.

This study demonstrates that increasing the number of clusters for modeling improves the ability to predict conflict as long as the models are robust.

This study investigates the quantity of clusters used in conflict modeling, while previous research assumes a specific quantity before modeling.

Understanding diversity amongst potential beneficiaries of land redistribution is of critical importance for both design and planning of successful land reform interventions. This study seeks to add to the existing literature on farming types, with specific emphasis on understanding diversity within a sub-group of commercially oriented or emerging smallholders.

Using a multivariate statistical analysis – principal component and cluster analyses applied to a sample of 442 commercially-oriented smallholders – five distinct clusters of emerging farmers are identified, using variables related to farmers' characteristics, income and expenditure and farm production indicators and willingness to participate in land redistribution. The five clusters are discussed in light of a predefined selection criteria that is based on the current policies and scholarly thinking.

The results suggest that there are distinct differences in farming types, and each identified cluster of farmers requires tailored support for the effective implementation of land reform. The identified homogenous sub-groups of smallholders, allows us to understand which farmers could be a better target for a successful land redistribution policy.

Most of the existing typology studies in South Africa tend to focus on general smallholders and in the Eastern Cape province; this study extends the literature by focussing on specific prime beneficiaries of land reform in three provinces. This study uses a more detailed dataset than the Statistics general and agricultural household surveys.