Mining the change of customer behavior in fuzzy time-interval sequential patterns

Comprehending changes of customer behavior is an essential problem that must be faced for survival in a fast-changing business environment. Particularly in the management of electronic commerce (EC), many companies have developed on-line shopping stores to serve customers and immediately collect buying logs in databases. This trend has led to the development of data-mining applications. Fuzzy time-interval sequential pattern mining is one type of serviceable data-mining technique that discovers customer behavioral patterns over time. To take a shopping example, (Bread, Short, Milk, Long, Jam), means that Bread is bought before Milk in a Short period, and Jam is bought after Milk in a Long period, where Short and Long are predetermined linguistic terms given by managers. This information shown in this example reveals more general and concise knowledge for managers, allowing them to make quick-response decisions, especially in business. However, no studies, to our knowledge, have yet to address the issue of changes in fuzzy time-interval sequential patterns. The fuzzy time-interval sequential pattern, (Bread, Short, Milk, Long, Jam), became available in last year; however, is not a trend this year, and has been substituted by (Bread, Short, Yogurt, Short, Jam). Without updating this knowledge, managers might map out inappropriate marketing plans for products or services and dated inventory strategies with respect to time-intervals. To deal with this problem, we propose a novel change mining model, MineFuzzChange, to detect the change in fuzzy time-interval sequential patterns. Using a brick-and-mortar transactional dataset collected from a retail chain in Taiwan and a B2C EC dataset, experiments are carried out to evaluate the proposed model. We empirically demonstrate how the model helps managers to understand the changing behaviors of their customers and to formulate timely marketing and inventory strategies.     

Microstructure Evolution and Tensile Properties of a Selectively Laser Melted CoNi-Base Superalloy

Metallurgical and Materials Transactions A - A high $$\gamma ^{\prime }$$ volume fraction CoNi-base superalloy with roughly equal amounts of cobalt and nickel was successfully processed through...     

The multivariable model-based control of the non-alumina electrolyte variables in aluminum smelting cells

This paper investigates the application of multivariable model-based control to improve the regulatory control of electrolyte temperature, aluminum fluoride concentration, liquidus temperature, superheat, and electrolyte height. Also examined are therappropriateness of different control structures and the possible inclusion of recently developed sensors for alumina concentration and individual cell duct flowrate, temperature, and heat loss. For the smelter in this study, the maximum improvement possible with a multivariable model-based controller is predicted to be 30–40% reduction in standard deviation in electrolyte temperature, aluminum fluoride concentration, liquidus temperature, and superheat, and around half this for electrolyte height. Three control structures were found to be appropriate; all are different than the existing control structure, which was found to be suboptimal. Linear Quadratic Gaussian controllers were designed for each control structure and their predicted performance compared. Comalco Research.     

Battery State-Of-Charge Based Altitude Controller for Small, Low Cost Multirotor Unmanned Aerial Vehicles

Small unmanned aerial vehicles (UAV) are typically driven by Lithium polymer batteries. The batteries have their own dynamics, which changes during discharge. Classical approaches to altitude control assume a time-invariant system and therefore fail. Adaptive controllers require an identified system model which is often unavailable. Battery dynamics can be characterized and used for a battery model-based controller. This controller is useful in situations when no feedback from actuators (such as RPM or thrust) is available. After measuring the battery dynamics for two distinct types of batteries, a controller is designed and experimentally verified, showing a consistent performance during entire discharge test and a consequent flight verification.     

Short-term prediction of traffic flow using a binary neural network

This paper introduces a binary neural network-based prediction algorithm incorporating both spatial and temporal characteristics into the prediction process. The algorithm is used to predict short-term traffic flow by combining information from multiple traffic sensors (spatial lag) and time series prediction (temporal lag). It extends previously developed Advanced Uncertain Reasoning Architecture (AURA) k-nearest neighbour (k-NN) techniques. Our task was to produce a fast and accurate traffic flow predictor. The AURA k-NN predictor is comparable to other machine learning techniques with respect to recall accuracy but is able to train and predict rapidly. We incorporated consistency evaluations to determine whether the AURA k-NN has an ideal algorithmic configuration or an ideal data configuration or whether the settings needed to be varied for each data set. The results agree with previous research in that settings must be bespoke for each data set. This configuration process requires rapid and scalable learning to allow the predictor to be set-up for new data. The fast processing abilities of the AURA k-NN ensure this combinatorial optimisation will be computationally feasible for real-world applications. We intend to use the predictor to proactively manage traffic by predicting traffic volumes to anticipate traffic network problems.     

Vinyl bottle performance with the addition of vinyl bottle flake

The effect of adding recycled vinyl bottle flake to virgin bottle compound has been studied. Vinyl bottles contains 10 to 50% recycled vinyl bottle flake were produced on a Bekum extrusion blow molding machine. The performance properties of the vinyl bottles were evaluated, and the optimum level of recycled flake was determined.     

Moving people tracking with detection by latent semantic analysis for visual surveillance applications

The latent semantic analysis (LSA) has been widely used in the fields of computer vision and pattern recognition. Most of the existing works based on LSA focus on behavior recognition and motion classification. In the applications of visual surveillance, accurate tracking of the moving people in surveillance scenes, is regarded as one of the preliminary requirement for other tasks such as object recognition or segmentation. However, accurate tracking is extremely hard under challenging surveillance scenes where similarity among multiple objects or occlusion among multiple objects occurs. Usual temporal Markov chain based tracking algorithms suffer from the ‘tracking error accumulation problem’. The accumulated errors can finally make the tracking to drift from the target. To handle the problem of tracking drift, some authors have proposed the idea of using detection along with tracking as an effective solution. However, many of the critical issues still remain unsettled in these detection based tracking algorithms. In this paper, we propose a novel moving people tracking with detection based on (probabilistic) LSA. By employing a novel ‘twin-pipeline’ training framework to find the latent semantic topics of ‘moving people’, the proposed detection can effectively detect the interest points on moving people in different indoor and outdoor environments with camera motion. Since the detected interest points on different body parts can be used to locate the position of moving people more accurately, by combining the detection with incremental subspace learning based tracking, the proposed algorithms resolves the problem of tracking drift during each target appearance update process. In addition, due to the time independent processing mechanism of detection, the proposed method is also able to handle the error accumulation problem. The detection can calibrate the tracking errors during updating of each state of the tracking algorithm. Extensive, experiments on various surveillance environments using different benchmark datasets have proved the accuracy and robustness of the proposed tracking algorithm. Further, the experimental comparison results clearly show that the proposed tracking algorithm outperforms the well known tracking algorithms such as ISL, AMS and WSL algorithms. Furthermore, the speed performance of the proposed method is also satisfactory for realistic surveillance applications.     

An analysis of ball-and-race milling part II. The babcock E 1.7 mill

A mathematical model was developed for a ball-and-race mill based on specific rates of breakage and primary fragment distributions. The model includes internal classification of particles falling back into the race and external classification due to the built-on classifier. It was demonstrated that the normalized primary fragment distribution produced in a pilot-scale Babcock E-type mill of 17 in. race diameter was the same as in the Hardgrove laboratory test mill and that the specific rates of breakage varied with particle size in the same manner. Steady-state continuous tests on the pilot-scale mill showed that breakage rates depended on the rate of feed, since the mill pulled less power at low feed rates. This effect plus the residence time effect gave coarser product size distributions at low and high feed rates than at a medium feed rate. Model simulations based on parameters measured in the Hardgrove mill correctly predicted the product size distribution from the E-type mill.     

Studies on the response of ‘Tifblue’ and ‘Woodard’ rabbiteye blueberries to fertilizers

A field study was conducted to investigate the effects of three fertilizer placement methods and 11 fertilizer formulations on plant survival, economics of replanting, yield, and fruit size of two rabbiteye blueberry (Vaccinium ashei Reade) cultivars Tifblue and Woodard. Fertilizer rates were either mixed with the soil under plants (before planting), sidedressed, or half-rate was placed under plants with the other half sidedressed. Fertilizers at planting caused a total of 34% Woodard and 12% Tifblue plants to die as compared to 6% for Woodard and none for Tifblue when no fertilizer was applied. Fertilizer formulas 2-1-1 and 5-10-10 oxide ratios placed under Woodard plants caused 100% mortality whereas Tifblue fertilized with oxide ratios 2-2-2 and 5-10-10 under plants reached the maximum 50% mortality. Splitting fertilizer placement under the side of the plants resulted in less Woodard plant mortality (30%) than placing all the fertilizer under plants (54%). However, plant mortality for the side/under group was not different than sidedressing all fertilizer (23%). A planted hectare in this experiment consisted of 1852 Tifblue and 926 Woodard plants, providing one Woodard plant to pollinate two Tifblue plants. Based on the results obtained, the 5-10-10 fertilizer placed under plants would be expected to cause half the Tifblue (926 plants) and all the Woodard (926 plants) to be replanted. Plant costs alone was estimated to be $2315 with additional replanting costs of $93 for Tifblue and much higher for Woodard. Sidedressing 5-10-10 fertilizer eight weeks after planting would have prevented plant mortality. Placement of various fertilizers under Woodard or Tifblue plants resulted in a significant increase in plant mortality as compared to side placement of fertilizers. Yield loss due to plant mortality increased linearly with the N level of the fertilizers (ranging from 0 to 2) for both cultivars. Woodard fruit size was not affected by fertilizer placement but placing fertilizers under Tifblue plants produced smaller yields with larger fruit than sidedressed plants. Blueberry plants in a virgin acidic soil may experience no reduction in yield if fertilization is completely eliminated.     

Thermal expansion of an SiC particle-reinforced aluminum composite

Using a push-rod dilatometer, we measured between 76 and 390 K the thermal expansion of a particle-reinforced-composite wrought plate obtained by powdermetallurgy methods. The particles, 30% by volume, consisted of monocrystals of -SiC with sizes near 10 m. The matrix consisted of a 6061 aluminum alloy with original particle sizes near 20 m. We found the perpendicular thermal expansivity, ₃, higher by 26% than the in-plane thermal expansivity, ₁-₂. These values differ from a rule-of-mixture prediction by –3 and –23%, respectively. All three _i, components lie outside the Rosen-Hashin bounds. Levin's isotropic model agrees within 10% with the ₁-₂-₃ average. Both the anisotropy and the bounds violation result from microstructural nonhomogeneity arising from processing. Rosen and Hashin's transverse-isotropicsymmetry relationships account approximately for these effects by introducing the anisotropic elastic constants. Using neutron diffraction, we determined that the SiC particles are textureless.Paper presented at the Tenth International Thermal Expansion Symposium, June 6–7, 1989, Boulder, Colorado, U.S.A.