Pattern validation for MODIS image mining of burned area objects

An image mining method was applied to Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data to estimate the area burned by forest fires occurring in Galicia (Spain) between 4 August and 15 August 2006. Five different inputs were considered: post-fire near-infrared reflectance (NIR) band, post-fire Normalized Difference Vegetation Index (NDVI) image, pre-fire and post-fire NDVI difference image and 4-μm and 11-μm thermal bands. The proposed image mining method consists of three steps: a pre-classification step, applying kernel smoothing, based on the fast Fourier transform (FFT), a modelling step applying Gaussian distributions on individual grid cells with deviating values, and a thresholding step classifying the model into burned and unburned classes. Polygons collected in the field with Global Positioning System (GPS) measurements from a helicopter permitted validation of the burned area estimation. A Z-test based on the κ statistic compared the accuracy of this estimation with the accuracies achieved by traditional methods based both on spectral changes in reflectance after the fire and active fire detection. The results showed a significant improvement (7.5%) in the accuracy of the burned area estimation after kernel smoothing. Burned area estimation based on the smoothed difference image between pre-fire and post-fire NDVI image had the highest accuracy (κ = 0.72). We conclude that the image mining algorithm successfully extracted burned area objects and that extraction was best when smoothing was applied prior to classification. Image mining methods based on using the κ statistic thus provide a valuable validation procedure when selecting the optimal MODIS input image for estimating burned area objects.     

Analysis of gap-free chlorophyll-a data from MODIS in Arabian Sea,reconstructed using DINEOF

ABSTRACT

The chlorophyll-a concentration (chl-a), which is an index of phytoplankton pigment present in the oceans, is considered as a key indicator of health of marine ecosystems that could have direct effect on the human life. In this study, spatial and temporal variability of chl-a in the Arabian Sea (AS) is examined using reconstructed cloud-free ocean colour data for the period 2002–2015. Data Interpolating Empirical Orthogonal Function method is used to reconstruct the missing data. Subsequently, wavelet analysis is applied on the reconstructed data to assess the temporal variability in terms of seasonal, intra-seasonal, and interannual variability of chl-a in the AS. Wavelet analysis clearly depicted the low-frequency, stationary modes or approximation levels inferring the monthly, seasonal, and annual mean of the signal, while the high-frequency, non-stationary modes indicated the local abnormalities. From the analysis of gap-free data, the presence of biennial mode of variability in the northern AS chl-a is observed. The analysis further showed the existence of intra-seasonal oscillations in the northern AS during summer monsoon and single dominant peak during winter monsoon. Chl-a appeared to decline slightly during the entire study period across all the selected regions of the AS. Also, it is observed that chl-a in the northwestern region is highly dynamic than in the other regions of the AS.     

Monitoring and modelling of chlorophyll-a concentrations in rivers using a high-resolution satellite image: a case study in the Nakdong river,Korea

The feasibility of using remote-sensing data with high spatial resolution was assessed for monitoring and modelling of chlorophyll-a (chl-a) in river waters. Two-band and three-band reflectance models including the red-edge band were examined as spectral coefficients using a RapidEye image for river waters, where the scale is smaller and narrower than for ocean waters. A red?red-edge?NIR three-band model calculated by a cubic function explained 73% of variance in the estimated data using the relationship between spectral indices such as absorption coefficients obtained using the model and chl-a concentrations and performed better than the red?red-edge two-band. Chl-a concentrations were simulated by a one-dimensional water quality model, QUALKO2, and image-derived and measured chl-a concentrations were applied in the calibration step of simulation. The image-derived chl-a dataset showed more stable calibration throughout the study area and enhanced the results rather than measured data. It is expected that chl-a estimation techniques using high resolution satellite data, RapidEye, have the capability to support rapid and widespread water quality monitoring and modelling, when a field dataset is not large or precise enough to do it, but still requires the improvement of estimation accuracy.     

Evaluation of spatio-temporal variations in chlorophyll-a in Lake Naivasha,Kenya: remote-sensing approach

Restoration of the ecosystem services and functions of lakes requires an understanding of the turbidity dynamics in order to arrive at informed environmental management decisions. The understanding of the spatio-temporal dynamics of turbidity requires frequent monitoring of the turbidity components such as chlorophyll-a concentration. In this study, we explored the use of Moderate Resolution Imaging Spectroradiometer Aqua (MODIS-Aqua) satellite data in studying the spatio-temporal changes in chlorophyll-a concentration in Lake Naivasha, a turbid tropical system. The temporal trend of chlorophyll-a concentration over the study period in the lake was also evaluated. The temporal trend assessment was achieved through the removal of periodic seasonal interference using Seasonal-Trend decomposition based on the LOESS (Local Regression) procedure. The resultant chlorophyll-a concentration maps derived from MODIS-Aqua satellite data give an indication of the monthly spatial variation in chlorophyll-a concentration from 2002 to 2012. The results of regression analyses between satellite-derived chlorophyll-a and in situ measurements reveal a high level of precision, but with a measureable bias with the satellite underestimating actual in situ measurements (R² = 0.65, P < 0.001). Although the actual values of the chlorophyll-a concentrations are underestimated, the significant relationship between satellite-derived chlorophyll-a and in situ measurements provides reliable information for studying spatial variations and temporal trends. In 2009 and 2010, it was difficult to detect chlorophyll-a from the MODIS-Aqua imagery, and this coincided with a period of the lowest water levels in Lake Naivasha. An inverse relationship between de-seasoned water level and chlorophyll-a concentration was evident. This study shows that MODIS-Aqua satellite data provide useful information on the spatio-temporal variations in Lake Naivasha, which is useful in establishing general trends that are more difficult to determine through routine ground measurements.     

Spatiotemporal variability of snow cover on Tibet,China using MODIS remote-sensing data

ABSTRACT

Snow cover is an important component of the cryosphere, and the study on spatial and temporal variations of snow cover is essential for understanding the consequences and impacts of climate change and water resources management. In this study, the temporal variation of snow-covered area (SCA) and spatial variability of snow-cover frequency (SCF) on Tibet is analysed based on the Moderate Resolution Imaging Spectroradiometer (MODIS)/Terra snow cover product (MOD10A2) from 2000 to 2015, and relationships with main climate variables are investigated. Results are as follows: (1) there is a very weak decreasing trend in annual mean SCA, and a slight increasing trend in autumn and winter and a slight decreasing trend in spring and more robust decreasing trend in summer for SCA are found. (2) The temporal variation of SCA is negatively correlated with temperature, whereas it is little correlated with corresponding precipitation. (3) The general trend of spatial SCF variation on Tibet, predominated by snow-cover variations in spring and autumn, tends to decrease in spring while it tends to increase in autumn. (4) The spatial variability of SCF is attributed to snow-cover variations in autumn and spring, which is more obvious in higher latitudes in autumn while it is more noticeable in lower-latitude southeastern plateau in spring. (5) The regions with higher variability of snow cover are main pastoral land and more prone to snow-related disaster in Tibet, becoming key zone of snow-cover monitoring and disaster prevention and mitigation.     

Towards a better assessment of the ecological status of coastal waters using satellite-derived chlorophyll-a concentrations

The application of the new Water Framework Directive (WFD) of the European Union will require a dense and frequent monitoring of chlorophyll-a near the coast. Not counting the transitional water bodies located in the vicinity of estuaries, not less than seventy four coastal water bodies have to be monitored along the coast of the French Atlantic continental shelf and the English Channel. All the available data have to be gathered to implement a comprehensive monitoring scheme. To this purpose, we evaluate the capacity of ocean colour imagery to complete the conventional in situ data set collected in coastal networks. Satellite-derived chlorophyll-a concentration is obtained by the application of a coastal Look-Up-Table to water-leaving radiance of the Sea-viewing Wide Field Instrument Sensor (SeaWiFS) for the 1998–2004 period. Seven years of satellite-derived and in situ chlorophyll-a concentrations are compared at seven representative stations of different water bodies. These comparisons show that the satellite products are reliable in most of the situations studied and throughout the seasons. Then the satellite imagery is used to classify the coastal waters following the eutrophication risk criterion of the WFD. This classification is made according to the percentile-90 of chlorophyll-a calculated during the productive season, from March to October. Despite a lack of sensor coverage over a small fraction of the near shore waters, this work shows that the satellite monitoring can considerably ease the application of the WFD.     

Spatiotemporal feature mining algorithm based on multiple minimum supports of pattern growth in Internet of Things

The temporal and spatial characteristics of users are involved in most Internet of Things (IoT) applications. The spatial and temporal movement patterns of users are the most direct manifestation of the temporal and spatial characteristics. The user’s interests, activities, experience and other characteristics are reflected by mobile mode. In view of the low clustering efficiency of moving objects in convergent pattern mining in the IoT, a spatiotemporal feature mining algorithm based on multiple minimum supports of pattern growth is proposed. Based on the temporal characteristics of user trajectories, frequent and asynchronous periodic spatiotemporal movement patterns are mined. Firstly, the location sequence is modeled, and the time information is added to the model. Then, a mining algorithm of asynchronous periodic sequential pattern is adopted. The algorithm is based on multiple minimum supports of pattern growth. According to multiple minimum supports, the sequential pattern of asynchronous period is mined deeply and recursively. Finally, the proposed method is validated and evaluated by Gowalla dataset, in which the user characteristics are truly reflected. It is shown by the experimental results that the average pointwise mutual information (PWI) of the proposed algorithm reaches 0.93. And the algorithm is proved to be effective and accurate.

     

Estimation of chlorophyll-a concentration in coastal waters with HJ-1A HSI data using a three-band bio-optical model and validation

Accurate assessment of phytoplankton chlorophyll-a (chl-a) concentration in turbid waters by means of remote sensing is challenging because of the optical complexity of case 2 waters. We applied a bio-optical model of the form [R^–1(λ₁) – R^–1(λ₂)](λ₃), where R(λ_i) is the remote-sensing reflectance at wavelength λ_i, to estimate chl-a concentration in coastal waters. The objectives of this article are (1) to validate the three-band bio-optical model using a data set collected in coastal waters, (2) to evaluate the extent to which the three-band bio-optical model could be applied to the spectral radiometer (SR) ISI921VF-512T data and the hyperspectral imager (HSI) data on board the Chinese HJ-1A satellite, (3) to evaluate the application prospects of HJ-1A HSI data in case 2 waters chl-a concentration mapping. The three-band model was calibrated using three SR spectral bands (λ₁ = 664.9 nm, λ₂ = 706.54 nm, and λ₃ = 737.33 nm) and three HJ-1A HSI spectral bands (λ₁ = 637.725 nm, λ₂ = 711.495 nm, and λ₃ = 753.750 nm). We assessed the accuracy of chl-a prediction with 21 in situ sample plots. Chl-a predicted by SR data was strongly correlated with observed chl-a (R² = 0.93, root mean square error (RMSE) = 0.48 mg m^–3, coefficient of variation (CV) (RMSE/mean(chl-a_mea)) = 3.72%). Chl-a predicted by HJ-1A HSI data was also closely correlated with observed chl-a (R² = 0.78, RMSE = 0.45 mg m^–3, CV (RMSE/mean(chl-a_mea)) = 7.51%). These findings demonstrate that the HJ-1A HSI data are promising for quantitative monitoring of chl-a in coastal case-2 waters.     

Probabilities of monthly median chlorophyll-a concentrations in subarctic,temperate and subtropical lakes

High concentrations of chlorophyll-a (chl-a) during summer are by definition a common problem in eutrophicated lakes. Several models have been designed to predict chl-a concentrations but are unable to estimate the probability of predicted concentrations or concentration spans during subsequent months. Two different methods were developed to compute the probabilities of obtaining a certain chl-a concentration. One method is built on discrete Markov chains and the other method on a direct relationship between median chl-a concentrations from two months. Lake managers may use these methods to detect and counteract the risk of high chl-a concentrations and algal blooms during coming months. Both methods were evaluated and applied along different scenarios to detect the probability to exceed chl-a concentration in different coming months. The procedure of computing probabilities is strictly based on general statistics which means that neither method is constrained for chl-a but can also be used for other variables. A user-friendly software application was developed to facilitate and extend the use of these two methods.     

Fast pattern recognition using normalized grey-scale correlation in a pyramid image representation

The ability to quickly locate one or more instances of a model in a grey scale image is of importance to industry. The recognition/localization must be fast and accurate. In this paper we present an algorithm which incorporates normalized correlation into a pyramid image representation structure to perform fast recognition and localization. The algorithm employs an estimate of the gradient of the correlation surface to perform a steepest descent search. Test results are given detailing search time by target size, effect of rotation and scale changes on performance, and accuracy of the subpixel localization algorithm used in the algorithm. Finally, results are given for searches on real images with perspective distortion and the addition of Gaussian noise.     

Spatio-temporal dynamics of phytoplankton and primary production in Lake Tanganyika using a MODIS based bio-optical time series

Lake Tanganyika, the second largest freshwater ecosystem in Africa, is characterised by a significant heterogeneity in phytoplankton concentration linked to its particular hydrodynamics. To gather a proper understanding of primary production, it is necessary to consider spatial and temporal dynamics throughout the lake. In the present work, daily MODIS-AQUA satellite measurements were used to estimate chlorophyll-a concentrations and the diffuse attenuation coefficient (K490) for surface waters. The spatial regionalisation of Lake Tanganyika, based on Empirical Orthogonal Functions of the chlorophyll-a dataset (July 2002-November 2005), allowed for the separation of the lake in 11 spatially coherent and co-varying regions, with 2 delocalised coastal regions. Temporal patterns of chlorophyll-a showed significant differences between regions. Estimation of the daily primary production in each region indicates that the dry season is more productive than the wet season in all regions with few exceptions. Whole-lake daily primary productivity calculated on an annual basis (2003) was 646 ± 142 mg C m^− 2 day^− 1. Comparing our estimation to previous studies, photosynthetic production in Lake Tanganyika appears to be presently lower (about 15%), which is consistent with other studies which used phytoplankton biovolume and changes of δ¹³C in the lake sediments. The decrease in lake productivity in recent decades may be associated to changes in climate conditions.     

Long-term high-frequency validation of MODIS LST products in a Mediterranean mountain environment

The accuracy of Moderate-resolution Imaging Spectroradiometer (MODIS) level 3 1 km land surface temperature (LST) products was assessed through long-term validation carried out in a mountainous site in Sierra Nevada, southeast Spain. A total of 1458 day and night thermal images, acquired by Terra and Aqua satellites during 2008, were processed and compared to ground-truth data recorded at the meteorological station of Robledal de Cañar with a frequency of one measurement every 10 min. The purpose of this investigation was to understand whether MODIS LST data can be used as input for climate models to be constructed for mountainous environments. Several trends in the MODIS LST data were observed, including the underestimation of daytime values and the overestimation of night-time values. Although all the data sets (Terra and Aqua, diurnal and nocturnal) showed high correlation coefficients with ground measurements, only night values maintained a relatively high accuracy of approximately 2°C of annual average error. Factors that may cause errors in the MODIS LST data, like acquisition angle, cloud, and snow cover, were analysed without conclusive results. High accuracy levels, i.e. close to 1°C, similar to other validation studies carried out over simpler and much more homogenous land-cover types such as cultivated fields, have been achieved for night images acquired during the summer months, thus making these datasets reliable for their use in climatic models over mountainous regions.     

Evaluation of standard and regional satellite chlorophyll-a algorithms for moderate-resolution imaging spectroradiometer (MODIS) in the Bohai and Yellow Seas,China: a comparison of chlorophyll-a magnitude and seasonality

Retrieval of satellite remotely sensed chlorophyll-a (chl-a) concentrations in coastal regions such as the Bohai and Yellow Seas (BYS) is challenging due to their complex oceanic and atmospheric optical properties. The standard OC3M (ocean chl-a three-band algorithm for MODIS (moderate-resolution imaging spectroradiometer)) algorithm has been widely used in the BYS, despite well-known uncertainties about its accuracy in terms of absolute magnitude. This was based on the belief that OC3M chl-a is capable of representing reliable relative spatial and temporal patterns of sea surface chl-a concentrations. In this study, the ability of the standard OC3M chl-a algorithm to reproduce accurate seasonality patterns was evaluated, based on comparisons with in situ chl-a measurements in the BYS. The results quantified the overestimation by the standard OC3M algorithm with a median absolute percentage difference of 98.48% and a median relative difference of 1.13 mg m^?3.More importantly, the seasonality from OC3M chl-a was significantly biased relative to the seasonal patterns of in situ chl-a. In addition, a regional GAM (generalized additive model)-based satellite chl-a algorithm was evaluated and compared with OC3M chl-a. The results showed the GAM chl-a improved accuracy in both magnitude and seasonality when compared with in situ chl-a, relative to that from OC3M chl-a.     

Spatio-temporal distribution of chlorophyll-a concentration derived from MODIS satellite remote-sensing and in situ observations in Sanya Bay,China

This study examined satellite chlorophyll-a (chl-a) concentration and in situ observations in Sanya Bay (SYB). In situ observation of chl-a was conducted four times per year at 12 sampling stations in SYB from January 2004 to October 2008. Monthly satellite chl-a was derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) during 2000–2012. This study compared satellite chl-a values to in situ measurements in SYB. The two data sets match well in the whole region except for two estuaries. Results show that the average in situ chl-a was 1.49 mg m^?3 in SYB. Chl-a was relatively higher (>2 mg m^?3) and more variable in coastal areas, with a tendency to decrease offshore (<0.4 mg m^?3). The chl-a level in summer displayed obviously vertical stratification, with higher values at the bottom and lower values at the surface. Analysis of monthly mean chl-a showed that the highest level (>2 mg m^?3) appeared in December, with the lowest in March (<1 mg m^?3). The gradients are ranked winter, autumn, summer and spring. There was higher chl-a in autumn and winter, which may be associated with the stronger wind monsoon then. Annual mean chl-a from 2000 to 2012 varied from 1.17 to 2.05 mg m^?3, with the minimum in 2001 and the maximum in 2005. The chl-a level presented a roughly increasing tendency from 2000 to 2012, which may be related to the increasing nutrients associated with the development of tourism and fishery.     

Antecedents of open source software defects: A data mining approach to model formulation, validation and testing

This paper develops tests and validates a model for the antecedents of open source software (OSS) defects, using Data and Text Mining. The public archives of OSS projects are used to access historical data on over 5,000 active and mature OSS projects. Using domain knowledge and exploratory analysis, a wide range of variables is identified from the process, product, resource, and end-user characteristics of a project to ensure that the model is robust and considers all aspects of the system. Multiple Data Mining techniques are used to refine the model and data is enriched by the use of Text Mining for knowledge discovery from qualitative information. The study demonstrates the suitability of Data Mining and Text Mining for model building. Results indicate that project type, end-user activity, process quality, team size and project popularity have a significant impact on the defect density of operational OSS projects. Since many organizations, both for profit and not for profit, are beginning to use Open Source Software as an economic alternative to commercial software, these results can be used in the process of deciding what software can be reasonably maintained by an organization.     

Dust source identification using MODIS: A comparison of techniques applied to the Lake Eyre Basin, Australia

The impact of mineral aerosol (dust) in the Earth's system depends on particle characteristics which are initially determined by the terrestrial sources from which the sediments are entrained. Remote sensing is an established method for the detection and mapping of dust events, and has recently been used to identify dust source locations with varying degrees of success. This paper compares and evaluates five principal methods, using MODIS Level 1B and MODIS Level 2 aerosol data, to: (a) differentiate dust (mineral aerosol) from non-dust, and (2) determine the extent to which they enable the source of the dust to be discerned. The five MODIS L1B methods used here are: (1) un-processed false colour composite (FCC), (2) brightness temperature difference, (3) Ackerman's (1997: J.Geophys. Res., 102, 17069-17080) procedure, (4) Miller's (2003:Geophys. Res. Lett. 30, 20, art.no.2071) dust enhancement algorithm and (5) Roskovensky and Liou's (2005: Geophys. Res. Lett. 32, L12809) dust differentiation algorithm; the aerosol product is MODIS Deep Blue (Hsu et al., 2004: IEEE Trans. Geosci. Rem. Sensing, 42, 557-569), which is optimised for use over bright surfaces (i.e. deserts). These are applied to four significant dust events from the Lake Eyre Basin, Australia. OMI AI was also examined for each event to provide an independent assessment of dust presence and plume location. All of the techniques were successful in detecting dust when compared to FCCs, but the most effective technique for source determination varied from event to event depending on factors such as cloud cover, dust plume mineralogy and surface reflectance. Significantly, to optimise dust detection using the MODIS L1B approaches, the recommended dust/non-dust thresholds had to be considerably adjusted on an event by event basis. MODIS L2 aerosol data retrievals were also found to vary in quality significantly between events; being affected in particular by cloud masking difficulties. In general, we find that OMI AI and MODIS AQUA L1B and L2 data are complementary; the former are ideal for initial dust detection, the latter can be used to both identify plumes and sources at high spatial resolution. Overall, approaches using brightness temperature difference (BT10-11) are the most consistently reliable technique for dust source identification in the Lake Eyre Basin. One reason for this is that this enclosed basin contains multiple dust sources with contrasting geochemical signatures. In this instance, BTD data are not affected significantly by perturbations in dust mineralogy. However, the other algorithms tested (including MODIS Deep Blue) were all influenced by ground surface reflectance or dust mineralogy; making it impossible to use one single MODIS L1B or L2 data type for all events (or even for a single multiple-plume event). There is, however, considerable potential to exploit this anomaly, and to use dust detection algorithms to obtain information about dust mineralogy.     

Application of a generalized additive model (GAM) for estimating chlorophyll-a concentration from MODIS data in the Bohai and Yellow Seas,China

In optically complex waters, it is important to evaluate the accuracy of the standard satellite chlorophyll-a (chl-a) concentration algorithms, and to develop accurate algorithms for monitoring the dynamics of chl-a concentration. In this study, the Moderate Resolution Imaging Spectroradiometer (MODIS) satellite remote-sensing reflectance and concurrent in situ measured chl-a (2010–2013) were used to evaluate the standard OC3M algorithm (ocean chlorophyll-a three-band algorithm for MODIS) and Graver–Siegel–Maritorena model version 1 (GSM01) algorithm for estimating chl-a concentration in the Bohai and Yellow Seas (BYS). The results showed that the chl-a algorithms of OC3M and GSM01 with global default parameters presented poor performance in the BYS (the mean absolute percentage difference (MAPD) and coefficient of determination (R²) of OC3M are 222.27% and 0.25, respectively; the MAPD and R² of GSM01 are 118.08% and 0.07, respectively). A novel statistical algorithm based on the generalized additive model (GAM) was developed, with the aim of improving the satellite-derived chl-a accuracy. The GAM algorithm was established using the in situ measured chl-a concentration as the output variable, and the MODIS above water remote-sensing reflectance (visible bands at 412, 443, 469, 488, 531, 547, 555, 645, 667, and 678 nm) and bathymetry (water depth) as input variables. The MAPD and R² calculated between the GAM and the in situ chl-a concentration are 39.96% and 0.67, respectively. The results suggest that the GAM algorithm can yield a superior performance in deriving chl-a concentrations relative to the standard OC3M and GSM01 algorithms in the BYS.     

Post-launch analysis,calibration, and validation of CCD line of sight for a space-borne push-broom electro-optical sensor using a global image data set

This article describes a technique which estimates line-of-sight (LOS) parameters of each CCD array in a high-resolution electro-optical sensor on board a satellite using statistical information which is extracted automatically from a large number of real images acquired after launch. First, the focal length of panchromatic (PAN) CCD arrays is estimated by using thousands of ground control points (GCPs) converted to raw image space. Second, a model is introduced for deriving LOS parameters for multispectral (MS) CCD arrays in one focal plane assembly (FPA) such as the focal lengths and the first/last detector cell positions from automatically matched band-to-band (B2B) tie-points. Finally, the LOS parameters of all arrays in one FPA are updated using automatically matched tie-points in the overlap zone (OZ) in order to represent the geometrical relationship between the CCD arrays with the same spectral band in two FPAs. The proposed technique was applied to the calibration of a ground segment image processor for a currently operating high-resolution imaging satellite during its initial commissioning phase. This article describes the accuracy and robustness of the LOS parameters estimated by the proposed technique by using more than one hundred images covering full geographical locations and off-nadir tilt angle ranges.     

Defect spatial pattern recognition using a hybrid SOM–SVM approach in semiconductor manufacturing

As manufacturing geometries continue to shrink and circuit performance increases, fast fault detection and semiconductor yield improvement is of increasing concern. Circuits must be controlled to reduce parametric yield loss, and the resulting circuits tested to guarantee that they meet specifications. In this paper, a hybrid approach that integrates the Self-Organizing Map and Support Vector Machine for wafer bin map classification is proposed. The log odds ratio test is employed as a spatial clustering measurement preprocessor to distinguish between the systematic and random wafer bin map distribution. After the smoothing step is performed on the wafer bin map, features such as co-occurrence matrix and moment invariants are extracted. The wafer bin maps are then clustered with the Self-Organizing Map using the aforementioned features. The Support Vector Machine is then applied to classify the wafer bin maps to identify the manufacturing defects. The proposed method can transform a large number of wafer bin maps into a small group of specific failure patterns and thus shorten the time and scope for troubleshooting to yield improvement. Real data on over 3000 wafers were applied to the proposed approach. The experimental results show that our approach can obtain over 90% classification accuracy and outperform back-propagation neural network.     

Re-examination of interestingness measures in pattern mining: a unified framework

Numerous interestingness measures have been proposed in statistics and data mining to assess object relationships. This is especially important in recent studies of association or correlation pattern mining. However, it is still not clear whether there is any intrinsic relationship among many proposed measures, and which one is truly effective at gauging object relationships in large data sets. Recent studies have identified a critical property, null-(transaction) invariance, for measuring associations among events in large data sets, but many measures do not have this property. In this study, we re-examine a set of null-invariant interestingness measures and find that they can be expressed as the generalized mathematical mean, leading to a total ordering of them. Such a unified framework provides insights into the underlying philosophy of the measures and helps us understand and select the proper measure for different applications. Moreover, we propose a new measure called Imbalance Ratio to gauge the degree of skewness of a data set. We also discuss the efficient computation of interesting patterns of different null-invariant interestingness measures by proposing an algorithm, GAMiner, which complements previous studies. Experimental evaluation verifies the effectiveness of the unified framework and shows that GAMiner speeds up the state-of-the-art algorithm by an order of magnitude.