ENSEMBLE VOTING SYSTEM FOR MULTICLASS PROTEIN FOLD RECOGNITION

An optical sensor was used to make digital recordings of wingbeat waveforms for the five most common aphids found on Guam: Aphis craccivora Koch, A. gossypii Glover, A. nerii Fonscolombe, Pentalonia nigronervosa Coquerel, and Toxoptera citricida (Kirkaldy). Wingbeat frequencies for each species overlapped all other species. However, mean wingbeat frequencies were significantly different for all species. Wingbeat frequencies and harmonic patterns were extracted from the recordings and submitted to cluster analysis, which failed to separate species completely. Several nearest neighbor and probabilistic neural network classifiers were built using time series, frequency spectra, wingbeat frequencies, and harmonic patterns as input variables. These classifiers were evaluated by having them identify wingbeat waveforms from aphids collected and recorded after their construction. The best performing classifier model was a probabilistic artificial neural network trained using 256-bin frequency spectra as input. Sixty-nine percent of the waveforms presented to this network were identified correctly. This study demonstrates the feasibility of developing an insect flight monitor that automatically counts and identifies individual flying insects. Essential components of the monitoring system are a photosensor, a multimedia personal computer, and software that identifies wingbeat frequency spectra using an artificial neural network.

SCOPe (Structural Classification of Proteins—extended, http://scop.berkeley.edu) is a database of relationships between protein structures that extends the Structural Classification of Proteins (SCOP) database. SCOP is an expert-curated ordering of domains from the majority of proteins of known structure in a hierarchy according to structural and evolutionary relationships. SCOPe classifies the majority of protein structures released since SCOP development concluded in 2009, using a combination of manual curation and highly precise automated tools, aiming to have the same accuracy as fully hand-curated SCOP releases. SCOPe also incorporates and updates the ASTRAL compendium, which provides several databases and tools to aid in the analysis of the sequences and structures of proteins classified in SCOPe. SCOPe continues high-quality manual classification of new superfamilies, a key feature of SCOP. Artifacts such as expression tags are now separated into their own class, in order to distinguish them from the homology-based annotations in the remainder of the SCOPe hierarchy. SCOPe 2.06 contains 77,439 Protein Data Bank entries, double the 38,221 structures classified in SCOP.

High-throughput genome sequencing projects continue to churn out enormous amounts of raw sequence data. However, most of this raw sequence data is unannotated and, hence, not very useful. Among the various approaches to decipher the function of a protein, one is to determine its localization. Experimental approaches for proteome annotation including determination of a protein's subcellular localizations are very costly and labor intensive. Besides the available experimental methods, in silico methods present alternative approaches to accomplish this task. Here, we present two machine learning approaches for prediction of the subcellular localization of a protein from the primary sequence information. Two machine learning algorithms, k Nearest Neighbor (k-NN) and Probabilistic Neural Network (PNN) were used to classify an unknown protein into one of the 11 subcellular localizations. The final prediction is made on the basis of a consensus of the predictions made by two algorithms and a probability is assigned to it. The results indicate that the primary sequence derived features like amino acid composition, sequence order and physicochemical properties can be used to assign subcellular localization with a fair degree of accuracy. Moreover, with the enhanced accuracy of our approach and the definition of a prediction domain, this method can be used for proteome annotation in a high throughput manner. SubCellProt is available at www.databases.niper.ac.in/SubCellProt.

Passive polarization based imaging is a useful tool in computer vision and pattern recognition. A passive polarization imaging system forms a polarimetric image from the reflection of ambient light that contains useful information for computer vision tasks such as object detection (classification) and recognition. Applications of polarization based pattern recognition include material classification and automatic shape recognition. In this paper, we present two target detection algorithms for images captured by a passive polarimetric imaging system. The proposed detection algorithms are based on Bayesian decision theory. In these approaches, an object can belong to one of any given number classes and classification involves making decisions that minimize the average probability of making incorrect decisions. This minimum is achieved by assigning an object to the class that maximizes the a posteriori probability. Computing a posteriori probabilities requires estimates of class conditional probability density functions (likelihoods) and prior probabilities. A Probabilistic neural network (PNN), which is a nonparametric method that can compute Bayes optimal boundaries, and a -nearest neighbor (KNN) classifier, is used for density estimation and classification. The proposed algorithms are applied to polarimetric image data gathered in the laboratory with a liquid crystal-based system. The experimental results validate the effectiveness of the above algorithms for target detection from polarimetric data.

In machine learning applications, the reliability of predictions is significant for assisted decision and risk control. As an effective framework to quantify the prediction reliability, conformal prediction (CP) was commonly developed with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$k$ </tex-math></inline-formula> -nearest neighbor, known as CPKNN. However, the conventional CPKNN suffers from high variance and bias, and long computational time as the feature dimensionality increases. To address these limitations, a new CP framework-CP with shrunken centroids (CPSC) is proposed. It regularizes the class centroids to attenuate the irrelevant features and shrink the sample space for predictions and reliability quantification. We have also compared the CPSC with CP based on SVM, LightGBM, RF on the herbal medicine dataset, with the electronic nose as a case and assessed them in two tasks: 1) off-line prediction: the training set was fixed, and the accuracy on the testing set was evaluated and 2) online prediction with data augmentation: CPKNN and CPSC filtered unlabeled data to augment the training data based on the prediction reliability, and the final accuracy of the testing set was compared. The result shows that CPSC significantly outperformed CPKNN in both two tasks: 1) CPSC reached a significantly higher accuracy with lower computation cost, and with the same credibility output, CPSC generally achieved higher accuracy and 2) the data augmentation process with CPSC robustly manifested a statistically significant improvement in prediction accuracy with different reliability thresholds, and the augmented data were more balanced in classes. This novel CPSC provides higher prediction accuracy and better reliability quantification, which can assist in reliable decision support.

Nowadays, available audio corpora are rapidly increasing from fast growing Internet and digitized libraries. How to effectively classify and retrieve such huge databases is a challenging task. Content based technology is studied to automatically classify audio into hierarchy classes. Based on a small set of features selected by the sequential forward selection (SFS) method from 87 extracted ones, four classifiers, namely nearest neighbor (NN), modified k-nearest neighbor (k-NN), Gaussian mixture model (GMM), and probabilistic neural network (PNN) are compared. Experiments were conducted on a common database and a more comprehensive database built by the authors. Finally, the PNN classifier combined with Euclidean distance measurement was chosen for audio retrieval, using query by example.

Theknearest neighbor classification (k‐NN) is a very simple and popular method for classification. However, it suffers from a major drawback, it assumes constant local class posterior probability. It is also highly dependent on and sensitive to the choice of the number of neighborsk. In addition, it severely lacks the desired probabilistic formulation. In this article, we propose a Bayesian adaptive nearest neighbor method (BANN) that can adaptively select the shape of the neighborhood and the number of neighborsk. The shape of the neighborhood is automatically selected according to the concentration of the data around each query point with the help of discriminants. The neighborhood size is not predetermined and is kept free using a prior distribution. Thus, we are able to make the model to select the appropriate neighborhood size. The model is fitted using Markov Chain Monte Carlo (MCMC), so we are not using exactly one neighborhood size but a mixture ofk. Our BANN model is highly flexible, determining any local pattern in the data‐generating process, and adapting it to give an improved prediction. We have applied our model on four simulated data sets with special structures and five real‐life benchmark data sets. Our proposed BANN method demonstrates substantial improvement overk‐NN and discriminant adaptive nearest neighbor (DANN) in all nine case studies. It also outperforms the probabilistic nearest neighbor (PNN) in most of the data analyses. Copyright © 2010 Wiley Periodicals, Inc. Statistical Analysis and Data Mining 3: 92‐105, 2010

Structural class of protein is important in understanding of folding patterns. Effective and reliable computational methods are needed for prediction of protein structural class. In this paper, a novel method for prediction of protein structural class was proposed, which combined protein sequence information and predicted secondary structural feature, and used support vector machine classifier to classify attributes of protein. Jackknife cross-validation was taken to evaluate the the performance of proposed method, using three benchmark datasets. Results demonstrate that the proposed method combining the predicted secondary structural feature with sequence information is more efficient than the existing methods, which indicates the necessity to extract more information to improve protein structural class prediction.

The Structural Classification of Proteins (SCOP) database provides a detailed and comprehensive description of the relationships of all known protein structures. The classification is on hierarchical levels: the first two levels, family and superfamily, describe near and far evolutionary relationships; the third, fold, describes geometrical relationships. The distinction between evolutionary relationships and those that arise from the physics and chemistry of proteins is a feature that is unique to this database, so far. The database can be used as a source of data to calibrate sequence search algorithms and for the generation of population statistics on protein structures. The database and its associated files are freely accessible from a number of WWW sites mirrored from URL http://scop. mrc-lmb.cam.ac.uk/scop/.

In this study, I develop 10 alternative classification models using logit analysis, discriminant analysis, support vector machines, artificial neural networks, probabilistic neural networks, nearest neighbours, UTADIS and MHDIS for the detection of falsified financial statements. The models are developed using financial and nonfinancial data. The sample includes 398 financial statements, half of which were assigned a qualified audit opinion. I compare these alternatives methods using out‐of‐time and out‐of‐sample tests. The results are used to derive conclusions on the performance of the methods and to investigate the potential of developing models that will assist auditors in identifying fraudulent financial statements. Copyright © 2009 John Wiley &amp; Sons, Ltd.

The state of a meeting room (or of a classroom) provides important context to the level of interest (or participation level) in a meeting. For example, in one uninteresting meeting only one presenter presents text-filled slides, while other attendees are either checking their emails or even napping. In contrast, during a very interesting talk, all attendees are excited by the presentation and are involved in the discussions with frequent hand clapping. Thus, automatically detecting the state of a meeting room (or of a classroom) is an important and interesting problem. In this paper, we make the observation that the aforementioned group behaviors will have different motion signatures, as captured by a present Wi-Fi signal. Consequently, in this paper, we present a meeting room state detection approach leveraging environmental Wi-Fi signature. The Wi-Fi signal is provided by the existing access point (AP) in the meeting room and it is captured by one any Wi-Fi enabled device. The existing infrastructure provides all the information needed and no extra devices are needed. In our solution, different features are extracted from the raw Wi-Fi Received Signal Strength Indicator (RSSI) signal and four popular machine learning algorithms (e.g. support vector machine (SVM), decision tree, nearest neighbor and naive Bayes networks) are evaluated for detecting the state of a meeting room. Our approach is evaluated for two scenarios: classroom state detection and meeting room state detection. The experimental results show an accuracy of our proposed solution of 90.7% and 94.1% for classroom state detection and meeting/lab room state detection, respectively.

Closure to “Soil Water Retention Characteristics of Vertisols and Pedotransfer Functions Based on Nearest Neighbor and Neural Networks Approaches to Estimate AWC” by N. G. Patil, D. K. Pal, C. Mandal, and D. K. Mandal

In India, the oral cancers are usually presented in advanced stage of malignancy. It is critical to ascertain the diagnosis in order to initiate most advantageous treatment of the suspicious lesions. The main hurdle in appropriate treatment and control of oral cancer is identification and risk assessment of early disease in the community in a cost-effective fashion. The objective of this research is to design a data mining model using probabilistic neural network and general regression neural network (PNN/GRNN) for early detection and prevention of oral malignancy. The model is built using the oral cancer database which has 35 attributes and 1025 records. All the attributes pertaining to clinical symptoms and history are considered to classify malignant and non-malignant cases. Subsequently, the model attempts to predict particular type of cancer, its stage and extent with the help of attributes pertaining to symptoms, gross examination and investigations. Also, the model envisages anticipating the survivability of a patient on the basis of treatment and follow-up details. Finally, the performance of the PNN/GRNN model is compared with that of other classification models. The classification accuracy of PNN/GRNN model is 80% and hence is better for early detection and prevention of the oral cancer.

A very important part of machine learning is classification. One of the most important classification algorithms is the k-nearest neighbour algorithm. This algorithm works on the principle that a point belongs to the same class as the majority of its closest neighbours. In this approach, a point qualifies as one of the closest neighbours of another point based on the parameter ‘k’. We determine whether a point has a bearing on another neighbour’s class if it is one of the first ‘k’-nearest neighbours of the point in question. A major disadvantage of this algorithm is the fact that we are bound to take exactly ‘k’-nearest neighbours irrespective of the distance between a point and the neighbour. This is a major anomaly. We decide that we shall determine the importance of a point in an algorithm based on proximity to another point but do not pay any heed to the distance between those two points. Many solutions have been proposed to reduce the effect of this problem in the algorithm. This paper gets rid of this problem altogether and adds another dimension to the nearest neighbour classification model.

FSKNN: Multi-label text categorization based on fuzzy similarity and k nearest neighbors