Search results

The purpose of this paper is to present a fresh approach to stimulate individual creativity. It introduces a mathematical representation for creative ideas, six creativity operators and methods of matrix-algebra to evaluate, improve and stimulate creative ideas. Creativity begins with ideas to resolve a problem or tackle an opportunity. By definition, a creative idea must be simultaneously novel and useful. To inject analytic rigor into these concepts of creative ideas, the author introduces a feature-attribute matrix-construct to represent ideas, creativity operators that use ideas as operands and methods of matrix algebra. It is demonstrated that it is now possible to analytically and quantitatively evaluate the intensity of the variables that make an idea more, equal or less, creative than another. The six creativity operators are illustrated with detailed multi-disciplinary real-world examples. The mathematics and working principles of each creativity operator are discussed.

The unit of analysis is ideas, not theory. Ideas are man-made artifacts. They are represented by an original feature-attribute matrix construct. Using matrix algebra, idea matrices can be manipulated to improve their creative intensity, which are now quantitatively measurable. Unlike atoms and cute rabbits, creative ideas, do not occur in nature. Only people can conceive and develop creative ideas for embodiment in physical, non-physical forms, or in a mix of both. For example, as widgets, abstract theorems, business processes, symphonies, organization structures, and so on. The feature-attribute matrix construct is used to represent novelty and usefulness. The multiplicative product of these two matrices forms the creativity matrix. Six creativity operators and matrix algebra are introduced to stimulate and measure creative ideas. Creativity operators use idea matrices as operands. Uses of the six operators are demonstrated using multi-disciplinary real-world examples. Metrics for novelty, usefulness and creativity are in ratio scales, grounded on the Weber–Fechner Law. This law is about persons’ ability to discern differences in the intensity of stimuli.

Ideas are represented using feature-attribute matrices. This construct is used to represent novel, useful and creative ideas with more clarity and precision than before. Using matrices, it is shown how to unambiguously and clearly represent creative ideas endowed with novelty and usefulness. It is shown that using matrix algebra, on idea matrices, makes it possible to analyze multi-disciplinary, real-world cases of creative ideas, with clarity and discriminatory power, to uncover insights about novelty and usefulness. Idea-matrices and the methods of matrix algebra have strong explanatory and predictive power. Using of matrix algebra and eigenvalue analyses, of idea-matrices, it is demonstrated how to quantitatively rank ideas, features and attributes of creative ideas. Matrix methods operationalize and quantitatively measure creativity, novelty and usefulness. The specific elementary variables that characterize creativity, novelty and usefulness factors, can now be quantitatively ranked. Creativity, novelty and usefulness factors are not considered as monolithic, irreducible factors, vague “lumpy” qualitative factors, but as explicit sets of elementary, specific and measurable variables in ratio scales. This significantly improves the acuity and discriminatory power in the analyses of creative ideas. The feature-attribute matrix approach and its matrix operators are conceptually consistent and complementary with key extant theories engineering design and creativity.

First to define and specify ideas as feature-attribute matrices. It is demonstrated that creative ideas, novel ideas and useful ideas can be analytically and unambiguously specified and measured for creativity. It is significant that verbose qualitative narratives will no longer be the exclusive means to specify creative ideas. Rather, qualitative narratives will be used to complement the matrix specifications of creative ideas. First to specify six creativity operators enabling matrix algebra to operate on idea-matrices as operands to generate new ideas. This capability informs and guides a person’s intuition. The myth and dependency, on non-repeatable or non-reproducible serendipity, flashes of “eureka” moments or divine inspiration, can now be vacated. Though their existence cannot be ruled out. First to specify matrix algebra and eigen-value methods of quantitative analyses of feature-attribute matrices to rank the importance of elementary variables that characterize factors of novelty, usefulness and creativity. Use of verbose qualitative narratives of novelty, usefulness and creativity as monolithic “lumpy” factors can now be vacated. Such lumpy narratives risk being ambiguous, imprecise, unreliable and non-reproducible, Analytic and quantitative methods are more reliable and consistent. First to define and specify a method of “attacking the negatives” to systematically pinpoint the improvements of an idea’s novelty, usefulness and creativity. This procedure informs and methodically guides the improvements of deficient ideas.

The purpose of this paper is to improve the performance of E-nose in the detection of wound infection. Feature extraction and selection methods have a strong impact on the performance of pattern classification of electronic nose (E-nose). A new hybrid feature matrix construction method and multi-objective binary quantum-behaved particle swarm optimization (BQPSO) have been proposed for feature extraction and selection of sensor array.

A hybrid feature matrix constructed by maximum value and wavelet coefficients is proposed to realize feature extraction. Multi-objective BQPSO whose fitness function contains classification accuracy and a number of selected sensors is used for feature selection. Quantum-behaved particle swarm optimization (QPSO) is used for synchronization optimization of selected features and parameter of classifier. Radical basis function (RBF) network is used for classification.

E-nose obtains the highest classification accuracy when the maximum value and db 5 wavelet coefficients are extracted as the hybrid features and only six sensors are selected for classification. All results make it clear that the proposed method is an ideal feature extraction and selection method of E-nose in the detection of wound infection.

The innovative concept improves the performance of E-nose in wound monitoring, and is beneficial for realizing the clinical application of E-nose.

The purpose of this paper is to propose a new fault feature extraction scheme for the rolling element bearing.

The generalized Stockwell transform (GST) and the singular value ratio spectrum (SVRS) methods are combined. A time-frequency distribution measurement criterion named the energy concentration measurement (ECM) is initially used to determine the parameter of the optimal GST method. Then, the optimal GST is applied to conduct a time-frequency transformation for a raw signal. Subsequently, the two-dimensional time-frequency matrix is obtained. Finally, the improved singular value decomposition (SVD) analysis is used to conduct a noise reduction of the time-frequency matrix. The SVRS is proposed to select the effective singular values. Furthermore, the time-domain feature of the impact signal is obtained by taking the inverse GST transform.

The simulated and experimental signals are used to verify the superiority of the proposed method over conventional methods. The obtained results show that the proposed method can effectively extract fault features of the rolling element bearing.

This paper mainly discusses the application of GST and SVRS methods to analyze the weak fault feature extraction problem. The next research direction is to explore the application of the Hilbert Huang transform (HHT) and variational modal decomposition (VMD) in the impact feature extraction of rolling bearing.

In the present study, a new SVRS method is proposed to select the number of effective singular values. This paper proposed an effective way to obtain the fault feature in monitoring of rotating machinery.

More and more e-commerce web sites are using online customer reviews (OCRs) for customer segmentation. However, for durable products, customer purchases, and reviews only once for a long time, as while the product review score may highly affected by service factors or be “gently” evaluated. Existing regression or machine learning-based methods suffer from low accuracy when applied to the OCRs of durable products on e-commerce web sites. The purpose of this paper is to propose a new approach for customer segment analysis base on OCRs of durable products.

The research proposes a two-stage approach that employs latent class analysis (LCA): the feature-mention matrix construction stage and the LCA-based customer segmentation stage. The approach considers reviewers’ mention on product features, and the probability-based LCA method is adopted upon the characteristics of online reviews, to effectively cluster reviewers into specified segmentations.

The research finding is that, using feature-mention instead of feature-opinion records makes segment analysis more effective. The research also finds that, LCA method can better explain the characteristics of the OCR data of durable products for customer segmentation.

The research proposes a new approach to durable product review mining for customer segmentation analysis. The segment analysis result can provide supports for new product design and development, repositioning of existing products, marketing strategy development and product differentiation.

A new approach for customer segmentation analysis base on OCRs of durable products is proposed.

The purpose of this paper is to determine the suitability of adopting hybridized multicriteria decision-making models as a decision tool in engineering design. This decision tool will assist design engineers and manufacturers to determine a robust design concept before simulation and manufacturing while all the design features and sub features would have been identified during the decision-making process.

Fuzzy analytical hierarchy process (FAHP) and fuzzy technique for order preference by similarity to ideal solution (FTOPSIS) are hybridized and applied to obtain optimal design of a reconfigurable assembly fixture (RAF) from a set of alternative design concepts. Design features and sub features associated with the RAF are identified and compared using fuzzified pairwise comparison matrices to obtain weights of their relative importance in the optimal design. The FAHP obtained the fuzzy synthetic extent (FSE) values of the design features and sub features. The FSE values are used as weights of the design features and sub features in generating the decision matrix. FTOPSIS and FTOPSIS based on left and right scores were adopted to predict effects of the weights. Results were obtained for normalized and unnormalized weights of the design features and its effects on the relative closeness coefficients of the design alternatives.

The improved performance of the FTOPSIS based on left and right scores is due to the involvement of the left and right scores of weights of the design features in the computation of distances from positive and negative ideal solutions. Embedding the weights of the design features in the normalized decision matrix before estimating the distances of the design concepts from ideal solutions reduces the dependency of the closeness coefficients on the weights of the design features. This also decreases the difference in the final values of the design concepts. In essence, the weights of the design features have an impact in the closeness coefficient. There is reduction in the closeness coefficients of the design concepts due to normalization of the weights of the design features. However, normalizing the weights of the design features did not affect the variations in the final values of the design concept. As the final value of the design concepts can be influenced by the normalized weights of the design features, it can be implied that normalization of weights of the sub features will also affect the decision matrix. The study has been able to proof that hybridizing FAHP and FTOPSIS can produce effective results for decisions on optimal design by the application of FTOPSIS based on left and right scores rather than the general FTOPSIS.

This research develops a hybridized multicriteria decision-making model for decision-making in engineering design. It presents a detailed extension of hybridized FAHP and FTOPSIS based on left and right scores as a useful tool for considering the relative importance of design features and sub features in optimal design selection.

Text mining is growing in importance proportionate to the growth of unstructured data and its applications are increasing day by day from knowledge management to social media analysis. Mapping skillset of a candidate and requirements of job profile is crucial for conducting new recruitment as well as for performing internal task allocation in the organization. The automation in the process of selecting the candidates is essential to avoid bias or subjectivity, which may occur while shuffling through thousands of resumes and other informative documents. The system takes skillset in the form of documents to build the semantic space and then takes appraisals or resumes as input and suggests the persons appropriate to complete a task or job position and employees needing additional training. The purpose of this study is to extend the term-document matrix and achieve refined clusters to produce an improved recommendation. The study also focuses on achieving consistency in cluster quality in spite of increasing size of data set, to solve scalability issues.

In this study, a synset-based document matrix construction method is proposed where semantically similar terms are grouped to reduce the dimension curse. An automated Task Recommendation System is proposed comprising synset-based feature extraction, iterative semantic clustering and mapping based on semantic similarity.

The first step in knowledge extraction from the unstructured textual data is converting it into structured form either as Term frequency–Inverse document frequency (TF-IDF) matrix or synset-based TF-IDF. Once in structured form, a range of mining algorithms from classification to clustering can be applied. The algorithm gives a better feature vector representation and improved cluster quality. The synset-based grouping and feature extraction for resume data optimizes the candidate selection process by reducing entropy and error and by improving precision and scalability.

The productivity of any organization gets enhanced by assigning tasks to employees with a right set of skills. Efficient recruitment and task allocation can not only improve productivity but also cater to satisfy employee aspiration and identifying training requirements.

Industries can use the approach to support different processes related to human resource management such as promotions, recruitment and training and, thus, manage the talent pool.

The task recommender system creates knowledge by following the steps of the knowledge management cycle and this methodology can be adopted in other similar knowledge management applications.

The efficacy of the proposed approach and its enhancement is validated by carrying out experiments on the benchmarked dataset of resumes. The results are compared with existing techniques and show refined clusters. That is Absolute error is reduced by 30 per cent, precision is increased by 20 per cent and dimensions are lowered by 60 per cent than existing technique. Also, the proposed approach solves issue of scalability by producing improved recommendation for 1,000 resumes with reduced entropy.

Visual tracking technology enables industrial robots interacting with human beings intelligently. However, due to the complexity of the tracking problem, the accuracy of visual target tracking still has great space for improvement. This paper aims to propose an accurate visual target tracking method based on standard hedging and feature fusion.

For this study, the authors first learn the discriminative information between targets and similar objects in the histogram of oriented gradients by feature optimization method, and then use standard hedging algorithms to dynamically balance the weights between different feature optimization components. Moreover, they penalize the filter coefficients by incorporating spatial regularization coefficient and extend the Kernelized Correlation Filter for robust tracking. Finally, a model update mechanism to improve the effectiveness of the tracking is proposed.

Extensive experimental results demonstrate the superior performance of the proposed method comparing to the state-of-the-art tracking methods.

Improvements to existing visual target tracking algorithms are achieved through feature fusion and standard hedging algorithms to further improve the tracking accuracy of robots on targets in reality.

Lesson study (LS) is a collaborative form of action research. Collaboration is central to LS methodology, therefore exploring and expanding the understanding of the collaborative features that occur in LS is a priority. This paper explores the features of collaboration in existing publications on LS to consider if, as Quaresma (2020) notes, collaboration is simplistically referred to within LS research.

Utilising a qualitative review of LS literature to explore LS collaboration through Mynott's (2019) outcome model and Huxham and Vangen's (2005) theory of collaborative advantage and inertia. 396 publications using “lesson study” and “collaboration” as key words were considered and reviewed, with 26 articles further analysed and coded, generating a collaborative feature matrix.

While collaboration in LS is referred to generically in the articles analysed, the authors found examples where collaboration is considered at a meta, meso and micro level (Lemon and Salmons, 2021), and a balance between collaborative advantage and inertia. However, only a small proportion of LS publications discuss collaboration in depth and, while the matrix will support future research, more focus needs to be given to how collaboration functions within LS.

Through answering Robutti et al.'s (2016) question about what can be learnt from the existing LS research studies on collaboration, this paper builds on Mynott's (2019) outcome model by providing a detailed matrix of collaborative features that can be found in LS work. This matrix has applications beyond the paper for use by facilitators, leaders of LS, and researchers to explore their LS collaborations through improved understanding of collaboration.

Presents an inductive machine learning algorithm called CLILP2 that learns multiple covers for a concept from positive and negative examples. Although inductive learning is an error‐prone process, multiple meaning interpretation of the examples is utilized by CLILP2 to compensate for the narrowness of induction. The algorithm is tested on data sets representing three different domains. Analyses the complexity of the algorithm and compares the results with those obtained by others. Employs measures of specificity, sensitivity, and predictive accuracy which are not usually used in presenting machine learning results, and shows that they evaluate better the “correctness” of the learned concepts. The study is published in two parts: I – the CLILP2 algorithm; II – experimental results and conclusions.

Expert Cloud as a new class of cloud systems enables its users to request and share the skill, knowledge and expertise of people by employing internet infrastructures and cloud concepts. Since offering the most appropriate expertise to the customer is one of the clear objectives in Expert Cloud, colleague recommendation is a necessary part of it. So, the purpose of this paper is to develop a colleague recommender system for the Expert Cloud using features matrices of colleagues.

The new method is described in two phases. In the first phase, all possible colleagues of the user are found through the filtering mechanism and next features of the user and possible colleagues are calculated and collected in matrices. Six potential features of colleagues including reputation, expertise, trust, agility, cost and field of study were proposed. In the second phase, the final score is calculated for every possible colleague and then top-k colleagues are extracted among users. The survey was conducted using a simulation in MATLAB Software. Data were collected from Expert Cloud website. The method was tested using evaluating metrics such as precision, accuracy, incorrect recommendation and runtime.

The results of this study indicate that considering more features of colleagues has a positive impact on increasing the precision and accuracy of recommending new colleagues. Also, the proposed method has a better result in reducing incorrect recommendation.

In this paper, the colleague recommendation issue in the Expert Cloud is pointed out and the solution approach is applied into the Expert Cloud website.