Clustering retail products based on customer behaviour

The categorization of retail products is essential for the business decision-making process. It is a common practice to classify products based on their quantitative and qualitative characteristics. In this paper, we use a purely data-driven approach. Our clustering of products is based exclusively on the customer behaviour. We propose a method for clustering retail products using market basket data. Our model is formulated as an optimization problem which is solved by a genetic algorithm. It is demonstrated on simulated data how our method behaves in different settings. The application using real data from a Czech drugstore company shows that our method leads to similar results in comparison with the classification by experts. The number of clusters is a parameter of our algorithm. We demonstrate that if more clusters are allowed than the original number of categories is, the method yields additional information about the structure of the product categorization.     

Can we make silica luminescent?

It is still a challenge to develop optoelectronic or even photonic devices based on silicon technology. Silicon and its oxide, as well-known, do not possess direct electronic band transitions and, therefore, are not luminescent. The remaining weak light emission is based on intrinsic and extrinsic defect luminescence. Thus our investigations are extended to ion implantation, mainly to over-stoichiometric injection or isoelectronic substitution of the both constituents silicon or oxygen, i.e. by ions of the group IV (C, Si, Ge, Sn, Pb) or the group VI (O, S, Se). Such implantation produces new luminescence bands in silica layers, partially with optical electronic–vibronic transitions and respective multimodal spectra. In this context, special interest should be directed to low-dimension nanocluster formation in silica layers.     

A swarm-inspired projection algorithm

In this paper a new data projection algorithm which was inspired by the foraging behaviors of doves is proposed. We name the new data projection the swarm-inspired projection (SIP) algorithm. The algorithm allows us to visually estimate the number of clusters existing in a data set. Based on the projection result, we may then partition the data set into the corresponding number of clusters. The SIP algorithm regards each data pattern in a data set as a piece of crumb which will be sequentially tossed to a flock of doves on the ground. The doves will adjust their physical positions to compete for crumbs. Gradually, the flock of doves will be divided into several groups according to the distributions of the crumbs. The formed groups will naturally correspond to the underlying data structures in the data set. By viewing the scatter plot of the final positions of the doves we can estimate the number of clusters existing in the data set. Several data sets were used to demonstrate the effectiveness of the proposed SIP algorithm.     

On a Method for Detecting Clusters of Possible Uranium Deposits

When a two-dimensional map contains points that appear to be scattered somewhat at random, a question that often arises is whether groups of points that appear to cluster are merely exhibiting ordinary behavior that one can expect with any random distribution of points, or whether the clusters are too pronounced to be attributable to chance alone. In this paper a method for detecting clusters along a straight line.is applied to the two-dimensional map of bismuth-214 anomalies observed as part of the National Uranium Resource Evaluation Program in the Lubbock, Texas region. Some exact probabilities associated with this method are computed and compared with two approximate methods. The two methods for approximating probabilities work well in the cases examined, and can be used when it is prohibitive to obtain the exact probabilities.     

Genotypic variation of spring wheats for solvent retention capacities in relation to end-use quality

The solvent retention capacity (SRC) test is useful to measure flour components contributing to end-use functionality especially where the sample number is large and grain quantities are limited. The water, sodium carbonate, lactic acid, and sucrose SRC values ranged 78.0-98.0, 95.0-127.5, 101.5-139.0, and 125.0-163.0 g/100 g, respectively in 50 spring wheat varieties. Water SRC positively correlated with cookie spread ratio (r = 0.29) while Lactic acid SRC and Sucrose SRC positively correlated with cookie thickness (r = 0.31) and (r = 0.23). Negative correlation coefficients were observed in water SRC and cookie thickness (r = −0.27). Lactic acid SRC also negatively correlated with cookie spread ratio (r = −0.34). Sodium carbonate SRC was found to be negatively correlated with cookie diameter (r = −0.19) and cookie spread ratio (r = −0.16). The SRC test is a promising method for the evaluation of soft wheat varieties on the basis of their biochemical characteristics.     

Evaluation of downer reactor performance by catalytic ozone decomposition

Radial distribution profiles of ozone concentrations were measured along an 8.50 m high and 0.09 m inside diameter gas/solid co-current down-flow circulating fluidized bed (downer) to characterize the reactor performance. Tests were conducted under a series of operating conditions at room temperature and near atmospheric pressure, with FCC particles as the bed material. Results show that the concentration distribution of the ozone tracer gas correlates well with the flow structure of the downer. There is quite a uniform radial distribution of ozone concentrations in the core region of all tested axial sections in the fully developed region of the downer, except for the near-wall region where there is a sharp decrease in ozone concentration. And there exists a relatively significant non-uniform distribution in the entrance acceleration region of the downer.     

Preference segments: A deeper understanding of consumer acceptance or a serving order effect?

Cluster analysis is used widely to group consumers into homogenous “preference segments” from their patterns of overall liking ratings. In several case studies, cross tabulation of sample frequencies by serving orders reveals the assignment of consumers to segments may be due to the effect of serving order as opposed to genuine differences in acceptance patterns. This effect is defined as the “serving-order-by-segment effect”. To nullify this effect, four adjustment factors were developed and applied to the assessors’ data from six consumer tests. The adjustments differed in two ways: serving position and samples. Regarding serving position, two of the adjustments affected the first-serve responses only, while the other two adjustments affected the responses for all serving positions. Regarding samples, two of the adjustments used the same value to adjust the responses for all of the samples, while the other two adjustments used sample-specific values to adjust the responses for each sample. The number of samples in each of the six consumer tests varied from three to eleven. With the exception of the eleven sample study, cluster analysis on adjusted data yielded different conclusions than for non-adjusted data and results among the four adjustments varied. Serving-order-by-sample interaction plots may be useful in determining the most appropriate adjustment factor.     

Brittle fracture of material surface under low-energy ion bombardment

A theoretical study is made on thermoelastic stress arising during interaction of a low-energy ion with a solid. The conditions are formulated that are necessary and sufficient for the brittle fracture and cluster sputtering of the material to occur. The sputtering coefficient is determined depending on the ion species and energy and the characteristics of material.     

Study of sol–gel transition in calcium alginate system by population balance model

The growth of calcium alginate (Ca-alginate) clusters and its sol–gel transition phenomena were simulated with a population balance model. The classical statistical concept proposed by Flory–Stockmayer has been adopted and incorporated by the factor of cyclic structures formation into the model. Ca-alginate gelation was described as a random cross-linking of multifunctional monochains. The resulting discretized kinetic equations were solved numerically by parallel computation. After demonstrating the accuracy of this model, the effect of the cyclic structure, the reaction rate constants and the physical properties of the reactants (i.e., alginate and gelling agent of Ca²⁺ ions) were examined. The present work clarifies the mechanisms for forming ionotropic Ca-alginate gels and presents the concentration ranges where the transition between the sol state and the gel state occurs. The model allows for the estimation of the relative magnitude of stepwise reaction rate constants and the prediction of the sol–gel transition boundary. Finally, a reasonable agreement was obtained between the experimental results and model predictions.     

Windows 2000集群技术应用

介绍了Windows 2000集群技术的基本概念、工作原理和具体的实施方案,以及在 具体实施中应该注意的一些问题。