Harnessing linked knowledge sources for topic classification in social media

Topic classification (TC) of short text messages offers an effective and fast way to reveal events happening around the world ranging from those related to Disaster (e.g. Sandy hurricane) to those related to Violence (e.g. Egypt revolution). Previous approaches to TC have mostly focused on exploiting individual knowledge sources (KS) (e.g. DBpedia or Freebase) without considering the graph structures that surround concepts present in KSs when detecting the topics of Tweets. In this paper we introduce a novel approach for harnessing such graph structures from multiple linked KSs, by: (i) building a conceptual representation of the KSs, (ii) leveraging contextual information about concepts by exploiting semantic concept graphs, and (iii) providing a principled way for the combination of KSs. Experiments evaluating our TC classifier in the context of Violence detection (VD) and Emergency Responses (ER) show promising results that significantly outperform various baseline models including an approach using a single KS without linked data and an approach using only Tweets. Copyright 2013 ACM.

Bagging model with cost sensitive analysis on diabetes data

Diabetes patients might suffer from an unhealthy life, long-term treatment and chronic complicated diseases. The decreasing hospitalization rate is a crucial problem for health care centers. This study combines the bagging method with base classifier decision tree and costs-sensitive analysis for diabetes patients' classification purpose. Real patients' data collected from a regional hospital in Thailand were analyzed. The relevance factors were selected and used to construct base classifier decision tree models to classify diabetes and non-diabetes patients. The bagging method was then applied to improve accuracy. Finally, asymmetric classification cost matrices were used to give more alternative models for diabetes data analysis.

On stopwords, filtering and data sparsity for sentiment analysis of Twitter

Sentiment classification over Twitter is usually affected by the noisy nature (abbreviations, irregular forms) of tweets data. A popular procedure to reduce the noise of textual data is to remove stopwords by using pre-compiled stopword lists or more sophisticated methods for dynamic stopword identification. However, the effectiveness of removing stopwords in the context of Twitter sentiment classification has been debated in the last few years. In this paper we investigate whether removing stopwords helps or hampers the effectiveness of Twitter sentiment classification methods. To this end, we apply six different stopword identification methods to Twitter data from six different datasets and observe how removing stopwords affects two well-known supervised sentiment classification methods. We assess the impact of removing stopwords by observing fluctuations on the level of data sparsity, the size of the classifier's feature space and its classification performance. Our results show that using pre-compiled lists of stopwords negatively impacts the performance of Twitter sentiment classification approaches. On the other hand, the dynamic generation of stopword lists, by removing those infrequent terms appearing only once in the corpus, appears to be the optimal method to maintaining a high classification performance while reducing the data sparsity and substantially shrinking the feature space

Statistical analysis and optimal classification of blood cell populations using Gaussian distributions

This dissertation develops a new figure of merit to measure the similarity (or dissimilarity) of Gaussian distributions through a novel concept that relates the Fisher distance to the percentage of data overlap. The derivations are expanded to provide a generalized mathematical platform for determining an optimal separating boundary of Gaussian distributions in multiple dimensions. Real-world data used for implementation and in carrying out feasibility studies were provided by Beckman-Coulter. It is noted that although the data used is flow cytometric in nature, the mathematics are general in their derivation to include other types of data as long as their statistical behavior approximate Gaussian distributions. ^ Because this new figure of merit is heavily based on the statistical nature of the data, a new filtering technique is introduced to accommodate for the accumulation process involved with histogram data. When data is accumulated into a frequency histogram, the data is inherently smoothed in a linear fashion, since an averaging effect is taking place as the histogram is generated. This new filtering scheme addresses data that is accumulated in the uneven resolution of the channels of the frequency histogram. ^ The qualitative interpretation of flow cytometric data is currently a time consuming and imprecise method for evaluating histogram data. This method offers a broader spectrum of capabilities in the analysis of histograms, since the figure of merit derived in this dissertation integrates within its mathematics both a measure of similarity and the percentage of overlap between the distributions under analysis. ^

Multidimensional classification and diagnosis of leukemia in flow cytometry

This research is to establish new optimization methods for pattern recognition and classification of different white blood cells in actual patient data to enhance the process of diagnosis. Beckman-Coulter Corporation supplied flow cytometry data of numerous patients that are used as training sets to exploit the different physiological characteristics of the different samples provided. The methods of Support Vector Machines (SVM) and Artificial Neural Networks (ANN) were used as promising pattern classification techniques to identify different white blood cell samples and provide information to medical doctors in the form of diagnostic references for the specific disease states, leukemia. The obtained results prove that when a neural network classifier is well configured and trained with cross-validation, it can perform better than support vector classifiers alone for this type of data. Furthermore, a new unsupervised learning algorithm---Density based Adaptive Window Clustering algorithm (DAWC) was designed to process large volumes of data for finding location of high data cluster in real-time. It reduces the computational load to ∼O(N) number of computations, and thus making the algorithm more attractive and faster than current hierarchical algorithms.

Statistical analysis and meta-analysis of microarray data

The microarray technology provides a high-throughput technique to study gene expression. Microarrays can help us diagnose different types of cancers, understand biological processes, assess host responses to drugs and pathogens, find markers for specific diseases, and much more. Microarray experiments generate large amounts of data. Thus, effective data processing and analysis are critical for making reliable inferences from the data. ^ The first part of dissertation addresses the problem of finding an optimal set of genes (biomarkers) to classify a set of samples as diseased or normal. Three statistical gene selection methods (GS, GS-NR, and GS-PCA) were developed to identify a set of genes that best differentiate between samples. A comparative study on different classification tools was performed and the best combinations of gene selection and classifiers for multi-class cancer classification were identified. For most of the benchmarking cancer data sets, the gene selection method proposed in this dissertation, GS, outperformed other gene selection methods. The classifiers based on Random Forests, neural network ensembles, and K-nearest neighbor (KNN) showed consistently god performance. A striking commonality among these classifiers is that they all use a committee-based approach, suggesting that ensemble classification methods are superior. ^ The same biological problem may be studied at different research labs and/or performed using different lab protocols or samples. In such situations, it is important to combine results from these efforts. The second part of the dissertation addresses the problem of pooling the results from different independent experiments to obtain improved results. Four statistical pooling techniques (Fisher inverse chi-square method, Logit method. Stouffer's Z transform method, and Liptak-Stouffer weighted Z-method) were investigated in this dissertation. These pooling techniques were applied to the problem of identifying cell cycle-regulated genes in two different yeast species. As a result, improved sets of cell cycle-regulated genes were identified. The last part of dissertation explores the effectiveness of wavelet data transforms for the task of clustering. Discrete wavelet transforms, with an appropriate choice of wavelet bases, were shown to be effective in producing clusters that were biologically more meaningful. ^

Forecasting Responses of Seagrass Distributions to Changing Water Quality Using Monitoring Data

Extensive data sets on water quality and seagrass distributions in Florida Bay have been assembled under complementary, but independent, monitoring programs. This paper presents the landscape-scale results from these monitoring programs and outlines a method for exploring the relationships between two such data sets. Seagrass species occurrence and abundance data were used to define eight benthic habitat classes from 677 sampling locations in Florida Bay. Water quality data from 28 monitoring stations spread across the Bay were used to construct a discriminant function model that assigned a probability of a given benthic habitat class occurring for a given combination of water quality variables. Mean salinity, salinity variability, the amount of light reaching the benthos, sediment depth, and mean nutrient concentrations were important predictor variables in the discriminant function model. Using a cross-validated classification scheme, this discriminant function identified the most likely benthic habitat type as the actual habitat type in most cases. The model predicted that the distribution of benthic habitat types in Florida Bay would likely change if water quality and water delivery were changed by human engineering of freshwater discharge from the Everglades. Specifically, an increase in the seasonal delivery of freshwater to Florida Bay should cause an expansion of seagrass beds dominated by Ruppia maritima and Halodule wrightii at the expense of the Thalassia testudinum-dominated community that now occurs in northeast Florida Bay. These statistical techniques should prove useful for predicting landscape-scale changes in community composition in diverse systems where communities are in quasi-equilibrium with environmental drivers.

Biogeochemical Classification of South Florida’s Estuarine and Coastal Waters of Tropical Seagrasses

South Florida’s watersheds have endured a century of urban and agricultural development and disruption of their hydrology. Spatial characterization of South Florida’s estuarine and coastal waters is important to Everglades’ restoration programs. We applied Factor Analysis and Hierarchical Clustering of water quality data in tandem to characterize and spatially subdivide South Florida’s coastal and estuarine waters. Segmentation rendered forty-four biogeochemically distinct water bodies whose spatial distribution is closely linked to geomorphology, circulation, benthic community pattern, and to water management. This segmentation has been adopted with minor changes by federal and state environmental agencies to derive numeric nutrient criteria.

Habitat Map for Lizard Island reef, Australia derived from a photo-transect survey field data collected in December 2011 and September/October 2012

An object based image analysis approach (OBIA) was used to create a habitat map of the Lizard Reef. Briefly, georeferenced dive and snorkel photo-transect surveys were conducted at different locations surrounding Lizard Island, Australia. For the surveys, a snorkeler or diver swam over the bottom at a depth of 1-2m in the lagoon, One Tree Beach and Research Station areas, and 7m depth in Watson's Bay, while taking photos of the benthos at a set height using a standard digital camera and towing a surface float GPS which was logging its track every five seconds. The camera lens provided a 1.0 m x 1.0 m footprint, at 0.5 m height above the benthos. Horizontal distance between photos was estimated by fin kicks, and corresponded to a surface distance of approximately 2.0 - 4.0 m. Approximation of coordinates of each benthic photo was done based on the photo timestamp and GPS coordinate time stamp, using GPS Photo Link Software (www.geospatialexperts.com). Coordinates of each photo were interpolated by finding the gps coordinates that were logged at a set time before and after the photo was captured. Dominant benthic or substrate cover type was assigned to each photo by placing 24 points random over each image using the Coral Point Count excel program (Kohler and Gill, 2006). Each point was then assigned a dominant cover type using a benthic cover type classification scheme containing nine first-level categories - seagrass high (>=70%), seagrass moderate (40-70%), seagrass low (<= 30%), coral, reef matrix, algae, rubble, rock and sand. Benthic cover composition summaries of each photo were generated automatically in CPCe. The resulting benthic cover data for each photo was linked to GPS coordinates, saved as an ArcMap point shapefile, and projected to Universal Transverse Mercator WGS84 Zone 56 South. The OBIA class assignment followed a hierarchical assignment based on membership rules with levels for "reef", "geomorphic zone" and "benthic community" (above).

Classification of Southern Ocean krill and icefish echoes using random forests

Acknowledgements The authors thank the crews, fishers, and scientists who conducted the various surveys from which data were obtained. This work was supported by the Government of South Georgia and South Sandwich Islands. Additional logistical support provided by The South Atlantic Environmental Research Institute, with thanks to Paul Brickle. PF receives funding from the MASTS pooling initiative (TheMarine Alliance for Science and Technology for Scotland), and their support is gratefully acknowledged. MASTS is funded by the Scottish Funding Council (grant reference HR09011) and contributing institutions. SF is funded by the Natural Environment Research Council, and data were provided from the British Antarctic Survey Ecosystems Long-term Monitoring and Surveys programme as part of the BAS Polar Science for Planet Earth Programme. The authors also thank the anonymous referees for their helpful suggestions on an earlier version of this manuscript.

Detection and Classification of Whale Acoustic Signals

This dissertation focuses on two vital challenges in relation to whale acoustic signals: detection and classification.

In detection, we evaluated the influence of the uncertain ocean environment on the spectrogram-based detector, and derived the likelihood ratio of the proposed Short Time Fourier Transform detector. Experimental results showed that the proposed detector outperforms detectors based on the spectrogram. The proposed detector is more sensitive to environmental changes because it includes phase information.

In classification, our focus is on finding a robust and sparse representation of whale vocalizations. Because whale vocalizations can be modeled as polynomial phase signals, we can represent the whale calls by their polynomial phase coefficients. In this dissertation, we used the Weyl transform to capture chirp rate information, and used a two dimensional feature set to represent whale vocalizations globally. Experimental results showed that our Weyl feature set outperforms chirplet coefficients and MFCC (Mel Frequency Cepstral Coefficients) when applied to our collected data.

Since whale vocalizations can be represented by polynomial phase coefficients, it is plausible that the signals lie on a manifold parameterized by these coefficients. We also studied the intrinsic structure of high dimensional whale data by exploiting its geometry. Experimental results showed that nonlinear mappings such as Laplacian Eigenmap and ISOMAP outperform linear mappings such as PCA and MDS, suggesting that the whale acoustic data is nonlinear.

We also explored deep learning algorithms on whale acoustic data. We built each layer as convolutions with either a PCA filter bank (PCANet) or a DCT filter bank (DCTNet). With the DCT filter bank, each layer has different a time-frequency scale representation, and from this, one can extract different physical information. Experimental results showed that our PCANet and DCTNet achieve high classification rate on the whale vocalization data set. The word error rate of the DCTNet feature is similar to the MFSC in speech recognition tasks, suggesting that the convolutional network is able to reveal acoustic content of speech signals.

Hyperspectral Image Classification with Nonlinear Methods

This thesis introduces two related lines of study on classification of hyperspectral images with nonlinear methods. First, it describes a quantitative and systematic evaluation, by the author, of each major component in a pipeline for classifying hyperspectral images (HSI) developed earlier in a joint collaboration [23]. The pipeline, with novel use of nonlinear classification methods, has reached beyond the state of the art in classification accuracy on commonly used benchmarking HSI data [6], [13]. More importantly, it provides a clutter map, with respect to a predetermined set of classes, toward the real application situations where the image pixels not necessarily fall into a predetermined set of classes to be identified, detected or classified with.

The particular components evaluated are a) band selection with band-wise entropy spread, b) feature transformation with spatial filters and spectral expansion with derivatives c) graph spectral transformation via locally linear embedding for dimension reduction, and d) statistical ensemble for clutter detection. The quantitative evaluation of the pipeline verifies that these components are indispensable to high-accuracy classification.

Secondly, the work extends the HSI classification pipeline with a single HSI data cube to multiple HSI data cubes. Each cube, with feature variation, is to be classified of multiple classes. The main challenge is deriving the cube-wise classification from pixel-wise classification. The thesis presents the initial attempt to circumvent it, and discuss the potential for further improvement.

Using Image Processing Methods to Improve the Detection of Buried Explosive Threats in GPR Data

Current state of the art techniques for landmine detection in ground penetrating radar (GPR) utilize statistical methods to identify characteristics of a landmine response. This research makes use of 2-D slices of data in which subsurface landmine responses have hyperbolic shapes. Various methods from the field of visual image processing are adapted to the 2-D GPR data, producing superior landmine detection results. This research goes on to develop a physics-based GPR augmentation method motivated by current advances in visual object detection. This GPR specific augmentation is used to mitigate issues caused by insufficient training sets. This work shows that augmentation improves detection performance under training conditions that are normally very difficult. Finally, this work introduces the use of convolutional neural networks as a method to learn feature extraction parameters. These learned convolutional features outperform hand-designed features in GPR detection tasks. This work presents a number of methods, both borrowed from and motivated by the substantial work in visual image processing. The methods developed and presented in this work show an improvement in overall detection performance and introduce a method to improve the robustness of statistical classification.

Land cover classification of Samoylov Island and Landsat subpixel water cover of Lena River Delta, Siberia, with links to ESRI grid files

Ignoring small-scale heterogeneities in Arctic land cover may bias estimates of water, heat and carbon fluxes in large-scale climate and ecosystem models. We investigated subpixel-scale heterogeneity in CHRIS/PROBA and Landsat-7 ETM+ satellite imagery over ice-wedge polygonal tundra in the Lena Delta of Siberia, and the associated implications for evapotranspiration (ET) estimation. Field measurements were combined with aerial and satellite data to link fine-scale (0.3 m resolution) with coarse-scale (upto 30 m resolution) land cover data. A large portion of the total wet tundra (80%) and water body area (30%) appeared in the form of patches less than 0.1 ha in size, which could not be resolved with satellite data. Wet tundra and small water bodies represented about half of the total ET in summer. Their contribution was reduced to 20% in fall, during which ET rates from dry tundra were highest instead. Inclusion of subpixel-scale water bodies increased the total water surface area of the Lena Delta from 13% to 20%. The actual land/water proportions within each composite satellite pixel was best captured with Landsat data using a statistical downscaling approach, which is recommended for reliable large-scale modelling of water, heat and carbon exchange from permafrost landscapes.

Permanent open water surfaces in central Siberia from ENVISAT ASAR wide swath data 2003/2004, link to shapefile

Permanent water bodies not only store dissolved CO2 but are essential for the maintenance of wetlands in their proximity. From the viewpoint of greenhouse gas (GHG) accounting wetland functions comprise sequestration of carbon under anaerobic conditions and methane release. The investigated area in central Siberia covers boreal and sub-arctic environments. Small inundated basins are abundant on the sub-arctic Taymir lowlands but also in parts of severe boreal climate where permafrost ice content is high and feature important freshwater ecosystems. Satellite radar imagery (ENVISAT ScanSAR), acquired in summer 2003 and 2004, has been used to derive open water surfaces with 150 m resolution, covering an area of approximately 3 Mkm**2. The open water surface maps were derived using a simple threshold-based classification method. The results were assessed with Russian forest inventory data, which includes detailed information about water bodies. The resulting classification has been further used to estimate the extent of tundra wetlands and to determine their importance for methane emissions. Tundra wetlands cover 7% (400,000 km**2) of the study region and methane emissions from hydromorphic soils are estimated to be 45,000 t/d for the Taymir peninsula.