Clustering in the presence of side information: a non-linear approach

2019 ◽  
Vol 12 (2) ◽  
pp. 292-314
Author(s):  
Ahmad Ali Abin
Keyword(s):  
Linear Method ◽  
Quadratic Function ◽  
Background Knowledge ◽  
Kernel Functions ◽  
Linear Functions ◽  
Quadratic Functions ◽  
Target Space ◽  
Constrained Clustering ◽  
Content Type ◽  
Non Linear

Purpose Constrained clustering is an important recent development in clustering literature. The goal of an algorithm in constrained clustering research is to improve the quality of clustering by making use of background knowledge. The purpose of this paper is to suggest a new perspective for constrained clustering, by finding an effective transformation of data into target space on the reference of background knowledge given in the form of pairwise must- and cannot-link constraints. Design/methodology/approach Most of existing methods in constrained clustering are limited to learn a distance metric or kernel matrix from the background knowledge while looking for transformation of data in target space. Unlike previous efforts, the author presents a non-linear method for constraint clustering, whose basic idea is to use different non-linear functions for each dimension in target space. Findings The outcome of the paper is a novel non-linear method for constrained clustering which uses different non-linear functions for each dimension in target space. The proposed method for a particular case is formulated and explained for quadratic functions. To reduce the number of optimization parameters, the proposed method is modified to relax the quadratic function and approximate it by a factorized version that is easier to solve. Experimental results on synthetic and real-world data demonstrate the efficacy of the proposed method. Originality/value This study proposes a new direction to the problem of constrained clustering by learning a non-linear transformation of data into target space without using kernel functions. This work will assist researchers to start development of new methods based on the proposed framework which will potentially provide them with new research topics.

Prediction of vernalisation in three Australian Vrn responsive wheats

2003 ◽  
Vol 54 (3) ◽  
pp. 283 ◽  
Author(s):  
L. D. J. Penrose ◽  
H. M. Rawson ◽  
M. Zajac
Keyword(s):  
Quadratic Function ◽  
Linear Interpolation ◽  
Development Stage ◽  
Linear Functions ◽  
Quadratic Functions ◽  
Published Data ◽  
Floral Initiation ◽  
Maximum Temperatures ◽  
Double Ridge ◽  
Interpolation Functions

This study sought to better estimate vernalisation in winter wheats, so that their early development and time of anthesis can be better predicted. For this, an accurate relationship between temperature and the effectiveness of vernalisation is required. Using previously published data, our study found that the relationship between temperature and effectiveness of vernalisation can be suitably described by a quadratic function. In contrast, most previous studies used linear interpolation functions to describe vernalising effectiveness. These consist of a series of linear functions of temperature over adjoining temperature ranges. An advantage of quadratic functions is that they allow effectiveness of vernalisation to be described in terms of underlying physiological processes, and require the estimation of fewer parameters to predict wheat development. Our study found the cardinal temperatures for vernalisation to be –3�C, 6.5�C, and 15.9�C, that is for the lower, optimum, and maximum temperatures respectively. To allow for different upper temperature limits for vernalisation, 2 quadratic temperature-vernalising effectiveness functions were used to predict accumulated daily vernalisation at 3 field sites. These predictions of daily vernalisation were compared with corresponding estimates produced with 3 previously proposed linear interpolation functions. Varying degrees of agreement were found between estimates produced by the 2 types of vernalising effectiveness functions. Equations that have been developed to predict floral initiation in winter wheats have not been previously evaluated in Australian field environments. These equations utilise the same underlying relationship between accumulated daily vernalisation and a measure of floral initiation, often the appearance of double ridges. Two of these equations were used to predict the appearance of double ridges for a field-grown Australian winter wheat, JF87%014. Neither equation could satisfactorily predict the timing of the double ridge development stage for this wheat, whatever vernalising effectiveness function was used to predict vernalisation in the field. Both equations had greatest difficulty in predicting the double ridge stage, in environments where vernalisation most delayed development. This finding suggests that equations currently predicting floral initiation in winter wheats do not utilise an accurate relationship between accumulated vernalisation and floral initiation. An alternative method of predicting anthesis in winter wheats is to predict final leaf number, but this approach has not been reliably applied in environments where vernalising temperatures vary.

scholarly journals Eggplant growth as affected by bovine manure and magnesium thermophosphate rates

2008 ◽  
Vol 65 (1) ◽  
pp. 77-86 ◽  
Author(s):  
Marinice Oliveira Cardoso ◽  
Walter Esfrain Pereira ◽  
Ademar Pereira de Oliveira ◽  
Adailson Pereira de Souza
Keyword(s):  
Leaf Area ◽  
Dry Matter ◽  
Quadratic Function ◽  
Potassium Sulfate ◽  
Linear Functions ◽  
Quadratic Functions ◽  
Positive Interaction ◽  
Relative Growth ◽  
High K ◽  
Cow Urine

Plant growth is influenced by nutrient availability. The objective of this research was to study, under greenhouse conditions, eggplant growth as affected by rates of bovine manure and magnesium thermophosphate (g kg-1 and mg kg-1, respectively), according to a "Box central composite" matrix: 4.15-259; 4.15-1509; 24.15-259; 24.15-1509; 0.0-884; 28.3-884; 14.15-0,0; 14.15-1768; 14.15-884. Potassium sulfate (170 mg kg-1) and 200 mL per pot of cow urine solution were applied four times, but the concentration of the last two applications (200 mL/H2O L) was twice of that of the first two. Additional treatments: magnesium thermophosphate without cow urine and triple superphosphate with urea, both with nutrient levels equivalent to the bovine manure, P2O5 and potassium sulfate to the combination 14.15-884. The experimental design consisted of randomized blocks with four replicates. Leaf area (LA) and LA ratio increased as quadratic functions with manure rates, with negative interaction for thermophosphate. Leaf dry matter mass (DMM) had an increasing quadratic function with rates for both fertilizers. The higher combined rates of both fertilizers resulted in the smallest specific leaf area, but also the highest values of shoot and root DMM, total DMM and, with positive interaction in relation to root shoot dry matter ratio. The relative growth rate in stem height, and also in diameter, increased with manure, according to quadratic and linear functions, respectively. The cow urine effect was, in general, lower than that of urea. The plant's overall growth was more influenced by manure. Root DMM and shoot DMM were greater with high K and P.

An integrated system based on fuzzy genetic algorithm and neural networks for stock price forecasting

2014 ◽  
Vol 31 (3) ◽  
pp. 281-292 ◽  
Author(s):  
Noraddin Mousazadeh Abbassi ◽  
Mohammad Ali Aghaei ◽  
Mahdi Moradzadeh Fard
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Stock Market ◽  
Price Index ◽  
Stock Exchange ◽  
Linear Method ◽  
Integrated System ◽  
Content Type ◽  
Non Linear ◽  
Fuzzy Genetic

Purpose – The aim of this research is to predict the total stock market index of the Tehran Stock Exchange, using the compound method of fuzzy genetics and neural network, in order for the active participants of the finance market as well as macro decision makers to be able to predict the market trend. Design/methodology/approach – First, the prediction was done by neural network, then the output weight of optimum neural network was taken as standard to repeat this prediction using the genetic algorithm, and then the extracted pattern from the neural network was stated through discernible rules using fuzzy theory. Findings – The main attention of this paper is investors and traders to achieve a method for predicting the stock market. Concerning the results of previous research, which confirms the relative superiority of non-linear models in price index prediction, an appropriate model has been offered in this research by compounding the non-linear method such as fuzzy genetics and neural network. The results indicate superiority of the designed system in predicting price index of the Tehran Stock Exchange. Originality/value – This paper states its originality and value by compounding the non-linear method issues pattern to predict stock market, to encourage further investigation by academics and practitioners in the field.

scholarly journals TOPOLOGY-GUIDED TESSELLATION OF QUADRATIC ElEMENTS

2009 ◽  
Vol 19 (02) ◽  
pp. 195-211 ◽  
Author(s):  
SCOTT E. DILLARD ◽  
VIJAY NATARAJAN ◽  
GUNTHER H. WEBER ◽  
VALERIO PASCUCCI ◽  
BERND HAMANN
Keyword(s):  
Level Sets ◽  
Piecewise Linear ◽  
Quadratic Function ◽  
Linear Functions ◽  
Quadratic Functions ◽  
Piecewise Linear Functions ◽  
Reeb Graph ◽  
Triangle Meshes ◽  
Triangular Patch ◽  
Piecewise Quadratic Functions

Topology-based methods have been successfully used for the analysis and visualization of piecewise-linear functions defined on triangle meshes. This paper describes a mechanism for extending these methods to piecewise-quadratic functions defined on triangulations of surfaces. Each triangular patch is tessellated into monotone regions, so that existing algorithms for computing topological representations of piecewise-linear functions may be applied directly to the piecewise-quadratic function. In particular, the tessellation is used for computing the Reeb graph, a topological data structure that provides a succinct representation of level sets of the function.

Pengembangan Media Pembelajaran Interaktif Berdasarkan Teori Apos pada Materi Fungsi Kuadrat

2018 ◽  
Vol 6 (1) ◽  
pp. 1-10
Author(s):  
Arie Gusman ◽  
Kamid Kamid ◽  
Syamsurizal Syamsurizal
Keyword(s):  
Learning Theory ◽  
Teaching And Learning ◽  
Quadratic Function ◽  
Test Group ◽  
Quadratic Functions ◽  
High School Mathematics ◽  
Constructivist Learning Theory ◽  
Average Value ◽  
Long Time ◽  
Action Process

Learning quadratic functions that had been performed by the majority of vocational school and high school mathematics teacher in Kuala Tungkal is still using conventional learning media. The use of conventional learning media is experiencing a lot of obstacles, such as: a fairly long time in describing the graph function, especially when analyzing some quadratic function graphs with various characteristics. APOS is one of the constructivist learning theory which states that students learn through several stages, namely: action – process – object – schema. And to integrate into media APOS writer adapting ADDIE development model. The effectiveness of the use of media-based learning theory APOS seen from the student activity sheet can be concluded more increased activity of students in the learning process. Study of the test results, students were able to meet the completeness criteria stipulated minimum is 75. With an average value of learning outcomes, namely 87.14. It can be seen from the students' responses on a test group of small and large groups where it is concluded that researchers develop learning media can be categorized as good / interesting in the teaching and learning of mathematics.

Solvent Polarity Studies of the Water+t-Butyl Alcohol and Water+t-Butylamine Binary Systems With the Solvatochromic Dyes Nile Red and Pyridinium-N-phenoxide Betaine, Refractometry and Permittivity Measurements

1994 ◽  
Vol 47 (9) ◽  
pp. 1771 ◽  
Author(s):  
PK Kipkemboi ◽  
AJ Easteal
Keyword(s):  
Binary Systems ◽  
Hydrophobic Interactions ◽  
Nile Red ◽  
Solvent Polarity ◽  
Butyl Alcohol ◽  
Linear Functions ◽  
Non Linear ◽  
Organic Cosolvent ◽  
Two Parameters ◽  
Polarity Parameters

The empirical solvent polarity parameters ENR and ET for the solvatochromic compounds Nile Red (1) and pyridinium-N-phenoxide betaine (2), respectively, have been determined as a function of composition for water+t -butyl alcohol and water+t-butylamine binary mixtures, over the whole composition range at 298 K. For both systems the two parameters vary with composition in a strongly non-linear fashion, and the polarity of the mixture decreases with increasing proportion of the organic cosolvent. The non-linear variation of the polarity parameters is attributed to water-cosolvent hydrophobic interactions at low cosolvent contents, and hydrogen-bonding interactions at higher cosolvent contents. Permittivity and refractive index have also been measured at 298 K for both systems, and both properties are strongly non-linear functions of composition.

Separate and/or simultaneous generations of non-linear functions inherent in non-ohmic resistivities

1976 ◽  
Vol 18 (1) ◽  
pp. 51-61
Author(s):  
Yasuhiro Kobayashi∗ ◽  
Masaaki Ohkita ◽  
Michio Inoue ◽  
Masao Nakamura
Keyword(s):  
Linear Functions ◽  
Non Linear

Heat and mass transfer effects on MHD non-Newtonian fluids flow through an inclined thermally-stratified porous medium

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Gladys Tharapatla ◽  
Pamula Rajakumari ◽  
Ramana G.V. Reddy
Keyword(s):  
Mass Transfer ◽  
Porous Medium ◽  
Velocity Profile ◽  
Heat And Mass Transfer ◽  
Newtonian Fluids ◽  
Transfer Effects ◽  
Content Type ◽  
Non Linear ◽  
Homotopy Analysis ◽  
Flow Through

Purpose This paper aims to analyze heat and mass transfer of magnetohydrodynamic (MHD) non-Newtonian fluids flow past an inclined thermally stratified porous plate using a numerical approach. Design/methodology/approach The flow equations are set up with the non-linear free convective term, thermal radiation, nanofluids and Soret–Dufour effects. Thus, the non-linear partial differential equations of the flow analysis were simplified by using similarity transformation to obtain non-linear coupled equations. The set of simplified equations are solved by using the spectral homotopy analysis method (SHAM) and the spectral relaxation method (SRM). SHAM uses the approach of Chebyshev pseudospectral alongside the homotopy analysis. The SRM uses the concept of Gauss-Seidel techniques to the linear system of equations. Findings Findings revealed that a large value of the non-linear convective parameters for both temperature and concentration increases the velocity profile. A large value of the Williamson term is detected to elevate the velocity plot, whereas the Casson parameter degenerates the velocity profile. The thermal radiation was found to elevate both velocity and temperature as its value increases. The imposed magnetic field was found to slow down the fluid velocity by originating the Lorentz force. Originality/value The novelty of this paper is to explore the heat and mass transfer effects on MHD non-Newtonian fluids flow through an inclined thermally-stratified porous medium. The model is formulated in an inclined plate and embedded in a thermally-stratified porous medium which to the best of the knowledge has not been explored before in literature. Two elegance spectral numerical techniques have been used in solving the modeled equations. Both SRM and SHAM were found to be accurate.

Adaptive efficiency of the Mexican Stock Exchange

2018 ◽  
Vol 31 (2) ◽  
pp. 326-335
Author(s):  
Esmeralda Brito-Cervantes ◽  
Semei Coronado ◽  
Manuel Morales-García ◽  
Omar Rojas
Keyword(s):  
Market Efficiency ◽  
Granger Causality ◽  
Latin American ◽  
Stock Exchange ◽  
Information Process ◽  
Causality Test ◽  
Content Type ◽  
Regulatory Policies ◽  
Adaptive Efficiency ◽  
Non Linear

Purpose The purpose of this paper is to analyse the adaptive market efficiency in the price–volume (P–V) relationship of the stocks listed in the Mexican Stock Exchange. The period under study goes from 1982 to 2015. In order to detect causality and, thus, determine adaptive efficiency in the market, one linear and two non-linear tests are applied. There are few papers in the literature that study the P–V relationship in Latin American markets; as such, this paper may be of interest and importance to financial academics and practitioners alike. Design/methodology/approach The Diks and Panchenko (DP) non-parametric Granger causality and the Brooks and Hinich (BH) cross-bicorrelation tests are applied. Findings Derived from the DP test, the findings show that there exists bi-directional non-linear Granger causality in 25.71 per cent of the firms studied, compared to 8 per cent when applying the linear Granger causality test. Therefore, there is evidence of weak-form efficiency in the market. From the BH test, evidence is shown of the adaptive market efficiency, since 71.42 per cent of firms exhibited some form of non-linear dependence in certain periods of time. With these results, the information process should be better studied for a greater comprehension of regulatory policies in the market and better decision-making tools for the investors. Originality/value This paper complements studies on the P–V relationship and efficiency in a Latin American market.

Non-linear stability analysis of micropolar fluid lubricated journal bearings with turbulent effect

2019 ◽  
Vol 71 (1) ◽  
pp. 31-39
Author(s):  
Subrata Das ◽  
Sisir Kumar Guha
Keyword(s):  
Stability Analysis ◽  
Linear Stability ◽  
Micropolar Fluid ◽  
Journal Bearing ◽  
Journal Bearings ◽  
Turbulent Regime ◽  
Content Type ◽  
Non Linear ◽  
The Stability ◽  
Bearing System

Purpose The purpose of this paper is to investigate the effect of turbulence on the stability characteristics of finite hydrodynamic journal bearing lubricated with micropolar fluid. Design/methodology/approach The non-dimensional transient Reynolds equation has been solved to obtain the non-dimensional pressure field which in turn used to obtain the load carrying capacity of the bearing. The second-order equations of motion applicable for journal bearing system have been solved using fourth-order Runge–Kutta method to obtain the stability characteristics. Findings It has been observed that turbulence has adverse effect on stability and the whirl ratio at laminar flow condition has the lowest value. Practical implications The paper provides the stability characteristics of the finite journal bearing lubricated with micropolar fluid operating in turbulent regime which is very common in practical applications. Originality/value Non-linear stability analysis of micropolar fluid lubricated journal bearing operating in turbulent regime has not been reported in literatures so far. This paper is an effort to address the problem of non-linear stability of journal bearings under micropolar lubrication with turbulent effect. The results obtained provide useful information for designing the journal bearing system for high speed applications.

Sign in / Sign up

Export Citation Format

Share Document