ABSTRACT In the early phases of a construction project, generating accurate and timely quotations is important for assessing feasibility. Delays or significant revisions in quotations can lead to project cancellations, resulting in lost business opportunities. To address this challenge, we propose a machine learning framework called Jointly Trained Automation of Explainable Construction Material Knowledge ( JACK ), developed in collaboration with a construction company to estimate material requirements. Our methodology begins with pre‐processing estimation data, where construction materials are categorised into high‐level types to facilitate more efficient learning. To support this process, open‐source synthetic data generators were developed to help clarify structural patterns for JACK , which employs a cascaded learning approach during training. The evaluation phase leverages joint training to enhance model efficiency and presents results across 207 construction projects. We also investigate the effects of dropout layers, regression trees and synthetic data augmentation on prediction accuracy. Finally, we compare JACK against traditional regression‐based methods using a separate project set, where it demonstrates competitive performance. Overall, JACK achieves low error rates across a range of material types, with performance gains largely attributed to the benefits of cascaded learning.

In computer vision, most data are captured in 2D formats, limiting spatial understanding in real-world applications. This presents a challenge for fields such as architecture, construction, and robotics, where interpreting spatial relationships from minimal visual input is increasingly essential. This survey reviews recent advancements in extracting 3D features from 2D imagery, a critical task in these domains, where spatial accuracy and object orientation directly impact performance. We focus on three core areas: (1) disposition estimation, determining object pose; (2) joint modeling, constructing skeletal representations; and (3) scene reconstruction, generating spatially accurate environments. Each category is evaluated based on input modalities, performance metrics, and code availability. By providing a unified overview of these techniques, this paper highlights their practical value in enabling 3D reasoning from conventional 2D data.

In the construction industry, a project typically begins with the creation of two-dimensional (2D) building plans, defining the client’s specifications. Using these plans, a digital three-dimensional (3D) model is developed to visualize the anticipated outcome and to verify the model’s alignment with the client’s expectations. The process of converting from 2D to 3D can become time-intensive if there is a need for modifications or if the project’s overall complexity is high. To enhance efficiency and accuracy, this research introduces an end-to-end framework referred to as BIRD which stands for B uilding I mage R econstruction and D imensioning. BIRD is capable of accepting five 2D perspective drawings of a building as inputs and generating a proportionate 3D model of the building envelope as an output. This is accomplished through the integration of multiple techniques that use convolutional neural networks to extract a refined set of line segments, identify measurements, and align each perspective with the floor plan drawing. The key contributions of this study includes: (1) a novel deep learning model designed for the identification of line segments in building plans; (2) novel algorithms that facilitate the generation of information required for 3D modeling; (3) an end-to-end framework for building reconstruction; and (4) novel performance metrics specifically tailored for the 2D to 3D conversion challenge. The practical application of this research was validated through the use of complete building plans provided by an industry partner. In summary, it was observed that BIRD demonstrated high accuracy in the creation of 3D visualizations, highlighting its real-world efficacy. • A framework is proposed to convert 2D building plans into 3D envelope models. • A novel dual-branch approach is used to extract line segments from building plans. • A novel algorithmic approach works to wrap line segments in three-dimensional spaces. • The framework is ran on building plans from an industry partner with high accuracy. • A preliminary user interface is described to highlight the practical use in industry.

User feedback is an important aspect of any social platform. The platform needs to adapt based on user preferences, eliminating the need for user input over time. This paper describes the experimentation with feedback received from users on their comfort level with temperatures in their work environment. The building energy management system will adapt itself so that users do not need to adjust the temperatures. This study addresses a number of issues with the user feedback modeling including managing a data imbalance with the help of fuzzy clustering, ability to transfer the user models between rooms and across different buildings, as well as, predicting a degree of discomfort. The experiments are based on user feedback collected from two commercial buildings.

The primary objective of a building energy management system is to ensure the comfort of its occupants. This paper reports experiments with a user feedback system for a central temperature controlled building. Individual room temperatures are further adjusted based on occupant feedback. The number of feedback points are limited creating a data imbalance. The paper describes how crisp and fuzzy clustering can be used to sample the data points when there is no feedback from the users. The results of the proposed sampling is compared with other sampling strategies. The goal of the research is to maximize the recall of feedback through machine learning.

This paper compares the semantically meaningful machine learning algorithms with the black box models. The machine learning models are applied to a real world wearable dataset for biometric identification of individuals. The semantically meaningful decision tree is compared with more accurate black-box models such as neural networks, random forest, and support vector machines. The paper further explores the possibility of using unsupervised learning that uses linear distances for separating the categories. Since the distance from the center is used to delineate the clusters, the centroids of the unsupervised clusters provide a semantic profile of the categories. The crisp K-means clustering is enhanced with evolutionary algorithms that primarily uses the distance from the center as the primary criteria, but nudges the clustering towards known classification using a semi-supervised penalty. Finally, the use of rough sets is shown to provide notable semantic information with the help of the three-way decision principle.

This paper explores the possibility of using distance based semi-supervised learning for creating lower and upper approximations of biometric signals. An evolutionary approach applied to both crisp and rough clustering optimizes both the within cluster scatter and the precision of the classification. The proposed approach is demonstrated through data collected from a wearable headband that recorded EEG brain signals. The brain signals are recorded for a number of participants performing various tasks. The approach identifies medoids that can best identify the participants. The evolutionary semi-supervised crisp and rough clustering is shown to favorably compare with the conventional unsupervised algorithms such as K-means.

Clustering groups objects based on their similarity using unsupervised learning. Clustering is an NP hard problem. A number of clustering algorithms use heuristics to create a reasonable grouping of objects. However, clustering schemes created by different heuristic algorithms do not always complet ely agree with each other. For example, an object may belong to different clusters for different algorithms. Therefore, researchers have proposed a number of clustering ensemble techniques to combine the clustering schemes from different algorithms. This paper proposes a Rough Set based ensemble method for preserving the inherent order in clustering. The proposal is demonstrated with the help of daily price patterns of commodities, which are grouped based on Black Scholes volatility index as well as the distribution of prices.

This paper reports the application of a possibility and rough set based clustering to semantically segmented real-world databases. The approach is an improved version of the well-known k-modes algorithm. It is a soft clustering method that clusters instances with uncertain categorical values to different clusters using their membership degrees. The possibility theory is used for dealing with uncertainty in the values of attributes and in the memberships of clusters. Rough sets are used to detect clusters with rough boundaries. We demonstrate the effectiveness of the proposed approach with the help of two real-world databases: a retail store or transactions data set and a mobile phone data set. The numeric values of attributes are segmented into semantically meaningful linguistic values using a novel discretization method. These linguistic values can lead to more natural interpretation of knowledge using possibilistic degrees. The possibilistic degrees describe our knowledge relative to the values of attributes (fully plausible to occur, may occur, or rejected) and identify the level of uncertainty in memberships to different clusters. In addition, our method deduces peripheral objects by calculating the approximate sets as defined in the rough set theory. The k-modes enhanced with rough set and possibility theories can provide semantically meaningful information for decision making to the store owners (retails data set) and telecommunication companies (mobile phone data set).

Users of online review sites can benefit from knowing the profiles of the businesses, as well as the profiles of reviewers who reviewed the businesses. This paper describes crisp and fuzzy metaclustering techniques to evolve two recursively defined clustering schemes of both businesses and reviewers in parallel using a real-world dataset supplied by yelp.com. The objective is to profile the businesses and reviewers by grouping them based on similar characteristics. The novelty of the proposed approach is in the fact that the representations of both businesses and reviewers change dynamically throughout the metaclustering process. A business is represented by static information obtained from the database and dynamic information obtained from the clustering of reviewers who reviewed the business. Similarly, the reviewer representation augments the static representation from the database with profiles of businesses who are reviewed by these reviewers. The resulting web-based service provides a facility for users to find similar businesses/reviewers based on the category of the business, rating, number of reviews, and number of check-ins. It also provides a succinct profile of a business or reviewer based on these factors so that the users can put the reviews in context. Since an object can belong to multiple clusters in fuzzy metaclustering, it is possible to absorb some of the extreme groups consisting of outliers in one of the mainstream clusters. As a result, the fuzzy metaclustering leads to more uniformly distributed and moderate profiles.

In this paper, we develop a new meta-clustering approach using possibility and rough set theories to handle imperfection in real-world retail datasets. Our proposal is a soft meta-clustering approach that provides a framework for handling uncertainty in the belonging of an object to different clusters. The soft meta-clustering approach is based on the k-modes algorithm devoted for categorical data. Possibility theory is used to represent the uncertainty between objects and clusters through possibilistic membership degrees. Rough set theory is applied to indicate clusters with rough boundaries. The meta-clustering consists of double clustering a retail dataset that contains customer and product data. The meta-clustering is improved by the application of the possibility and rough set theories. An initial clustering of the customer data is performed. Then, a second clustering of product data using the results of the first clustering is applied. The two clustering schemes then evolve iteratively affecting each other recursively. We detail our results and describe the structure of the final clusters of customers and products to prove the effectiveness of our proposal.

Information granules allow us to abstract real world objects using their relevant attributes. Level of abstraction allows us to create a network of information granules that are connected to each other through a real-world relationship. For example, the phone calls are connected to the origin and destination phone numbers. This paper describes the relationship between clustering schemes by propagating the information through the connection between granules. We focus on the hierarchical granular graphs, where phone numbers represent the higher level (coarser) granules that are connected to phone calls which are lower level (finer) granules. The paper studies the effect of transferring crisp and fuzzy clustering schemes through the granular hierarchy. The fuzzy clustering is shown to provide a better technique for such a transfer. Rough clustering derived from fuzzy clustering is shown to be a good instrument for comparing crisp and fuzzy clustering schemes.

Granular computing represents an object as an information granule. Traditionally the information is derived from the primary source of data by recording events such as transactions, phone calls, user sessions, security breaches, and car trips. Much of the early data mining techniques used information granules generated from primary data sources. Recent data mining techniques such as ensemble classifiers and stacked regression use secondary sources of data obtained from initial data mining activities. Typically, these techniques use preliminary applications of data mining techniques for initial knowledge discovery. The knowledge acquired from the preliminary data mining is then used for more refined analysis. Granular computing can enable us to develop a formal framework for incorporating information from both primary and secondary sources of data. This enhanced granular representation can help us develop integrated data mining techniques. This paper proposes a novel recursive meta-clustering algorithm to demonstrate the versatility of granular computing for developing integrated data mining techniques to exploit primary and secondary knowledge sources.

With the gaining popularity of rough clustering, soft computing research community is studying relationships between rough and fuzzy clustering as well as their relative advantages. Both rough and fuzzy clustering are less restrictive than convention

Fuzzy C-means (FCM) and Rough K-means (RKM) algorithms are two popular soft clustering algorithms that allow for overlapping clusters. The overlapping clusters can be useful in applications where restrictions imposed by crisp clustering that force assignment of every object to a unique cluster may not be practical. Likewise RKM and FCM, interval set representation of clusters would also generate overlapping clusters. We present and discuss the interval set K-means algorithm (IKM). This paper applies RKM, FCM and IKM algorithms for clustering web visits to an educational site. The experimental comparison highlights various features of these three soft computing algorithms.

In this paper, we apply classification system denoted Belief Rough Set Classifier (BRSC) based on the hybridization of belief functions and rough sets to learn decision rules from uncertain data consisting of web usage. The uncertainty appears only in decision attributes and is handled by the Transferable Belief Model (TBM), one interpretation of the belief function theory. The web usage mining dataset was obtained from an educational website, where the visits were clustered based on study patterns. Instead of using crisp assignment of a visit to one of the clusters, the study associated a belief that the visit will belong to that cluster. Due to the uncertainty existing in the chosen web usage mining database, the feature selection step used to construct our BRSC will be based on dynamic core approach which allows getting better performance in uncertain and large database. To judge the performance of our classifier, we choose three evaluation criteria: accuracy, size and time requirement. Besides, we compare the results with those obtained from a similar classifier, namely Belief Decision Tree (BDT).

Most of the business decisions are based on cost and benefit considerations. Data mining techniques that make it possible for the businesses to incorporate financial considerations will be moremeaningful to the decisionmakers. Decision theoretic framework has been helpful in providing a better understanding of classification models. This study describes a semi-supervised decision theoretic rough set model. The model is based on an extension of decision theoretic model proposed by Yao. The proposal is used to model financial cost/benefit scenarios for a promotional campaign in a real-world retail store.

Quality of clustering is an important issue in application of clustering techniques. Most traditional cluster validity indices are geometry-based cluster quality measures. This paper proposes a cluster validity index based on the decision-theoretic rough set model by considering various loss functions. Experiments with synthetic, standard, and real-world retail data show the usefulness of the proposed validity index for the evaluation of rough and crisp clustering. The measure is shown to help determine optimal number of clusters, as well as an important parameter called threshold in rough clustering. The experiments with a promotional campaign for the retail data illustrate the ability of the proposed measure to incorporate financial considerations in evaluating quality of a clustering scheme. This ability to deal with monetary values distinguishes the proposed decision-theoretic measure from other distance-based measures. The proposed validity index can also be extended for evaluating other clustering algorithms such as fuzzy clustering.

This paper provides a broad overview of logical and black box approaches to fuzzy and rough hybridization. The logical approaches include theoretical, supervised learning, feature selection, and unsupervised learning. The black box approaches consist of neural and evolutionary computing. Since both theories originated in the expert system domain, there are a number of research proposals that combine rough and fuzzy concepts in supervised learning. However, continuing developments of rough and fuzzy extensions to clustering, neurocomputing, and genetic algorithms make hybrid approaches in these areas a potentially rewarding research opportunity as well.

Researchers have modelled existing hyperlink structure using graph theory. Such graph theoretical models have been useful for efficient Web navigation, especially for designing Web crawlers and search engines. This paper describes how sequence analysis of Web page accesses can be used to create weighted directed graphs. The information contained in such graphs can be used to improve the Web surfing experience. The paper proposes a concept of neighbourhood modelled after neighbourhood systems and rough sets to evaluate the effectiveness of navigation provided by a given hyperlink structure and Web usage data. Furthermore, a modification of the minimum spanning tree algorithm is used to create a hyperlink structure that provides optimal navigation.

This paper introduces a new paradigm for avoiding vehicle collisions in intelligent transportation systems. The key feature of the concept is that we introduce interactive control of different vehicular sub-systems ultimately leading to interdependent maneuvering of vehicles in order to avoid collision encounters. A general scenario of motion of vehicular systems passing a common point at the same time is studied here. Using a fuzzy logic-based controller, collision avoidance maneuvers of the individual vehicular sub-systems are realized. An inter-vehicle communication (IVC) system that facilitates interactions between the vehicles is studied as a top tier above the individual vehicular controllers. The concept that is adopted here is that in order to control a general system with participatory sub-systems of vehicles, all the influencing members have to interactively work together in order to effectively negotiate to achieve an optimal result in avoiding collision encounters. While the interactive operations are being carried out, each vehicular sub-system may act as a 'master' one time and a 'slave' at another time. This switching between the states of 'master' and 'slave' should be done efficiently to better prevent a collision encounter. Even though above paradigm is proved only for two interactive sub-systems, it is emphasized that the concept can be extended to any number of vehicular sub-systems to realize effective collision avoidance maneuvers.

Time can play a crucial role in the analysis of web usage. Temporal data mining has been an active area of research. However, there is little work on the analysis of cluster memberships over time. Typical clustering operations in web mining involve finding natural groupings of web resources or web users. Changes in clusters can provide important clues about the changing nature of the usage of a web site, as well as changing loyalties of web users. This paper addresses two different types of temporal changes in cluster analysis. The changes in cluster compositions over time and changes in cluster memberships of individual web users. The paper also proposes the concept of temporal cluster migration matrices (TCMM). The proposed matrices are shown to be useful for analyzing the changing nature of a web site as well as changing patronages of individual web users. TCMM can be used as a visualization technique to study results obtained from temporal data mining, which can be more complex because of the additional time dimension.

WIC Canada promotes the collaboration between Canadian researchers on Web Intelligence, facilitates exchange with other WIC centers. WIC Canada researchers work on a diversity of WI related research areas: foundations of Web Intelligence, Web-based support systems, Bayesian networks, and IntelligentWeb Information Systems (IWIS). Their results have appeared in reputable journals and the annual IEEE/WIC/ACM International Conference onWeb Intelligence. In the next few years, WIC Canada will focus on promotingWeb Intelligence research, attracting new members, and forming new centres. The WIC Canada will play the role of coordinating those activities. WIC Canada needs your input for its growth. Your participation will be greatly appreciated. All questions and suggestions should be directed to our Co-ordinator, Dr. Jingtao Yao, at [email protected].

Temporal Web usage mining involves application of data mining techniques on temporal Web usage data to discover temporal patterns, which describe the temporal behavior of Web users. Clusters and associations in Web usage mining do not necessarily have crisp boundaries. We introduce the temporal Web usage mining of Web users on one educational Web site, using the adapted Kohonen SOM based on rough set properties [L. J. Pawan et al. (2002)].

Similar to traditional data mining, three important Web mining operations include clustering, association, and sequential analysis. Typical clustering operations in Web mining involve finding natural groupings of Web resources or Web users. Researchers have pointed out some important differences between clustering in conventional applications and clustering in Web mining. For example, the clusters and associations in Web mining do not necessarily have crisp boundaries. Moreover, due to a variety of reasons inherent in Web browsing and Web logging, the likelihood of bad or incomplete data is higher. As a result, researchers have studied the possibility of using fuzzy sets in Web mining clustering applications. The paper describes how rough set theory can also be used to develop clustering schemes for Web mining. The unsupervised classification described in the paper uses properties of rough sets along with genetic algorithms to represent clusters as interval sets. The paper also describes the design of an experiment including data collection and the clustering process. The experiment is used to create interval set representations of groups of Web visitors.

This paper describes rough genetic algorithms based on the notion of interval values. A gene in a rough genetic algorithm can be represented using an interval. The paper explains how this generalization facilitates development of new genetic operators and evaluation measures. Various operations of rough genetic algorithms are illustrated using a simple document retrieval example. An experiment involving classification of interval‐valued traffic patterns demonstrates the usefulness of rough genetic algorithms for real‐world applications. A discussion on further extensions of rough genetic algorithms to rough sets is also included.

This article examines basic issues of data mining using the theory of rough sets, which is a recent proposal for generalizing classical set theory. The Pawlak rough set model is based on the concept of an equivalence relation. Recent research has shown that a generalized rough set model need not be based on equivalence relation axioms. The Pawlak rough set model has been used for deriving deterministic as well as probabilistic rules from a complete database. This article demonstrates that a generalized rough set model can be used for generating rules from incomplete databases. These rules are based on plausibility functions proposed by Shafer. The article also discusses the importance of rule extraction from incomplete databases in data mining. © 1998 John Wiley & Sons, Inc.

Many intelligent systems employ numeric degrees of belief supplied by the users to make decisions. However, the users may have difficulties in expressing their belief in terms of numeric values. The authors present a method for generating belief functions from symbolic information such as the qualitative preference relationships. The method of generating belief functions provides a practical interface between the users and a decision support system. It can be argued that the ability to generate numeric judgments with nonnumeric inputs is essential in the development of approximate reasoning systems. The proposed method can provide an important component for these systems by transforming qualitative information into quantitative information.>