Towards automated content analysis of discussion transcripts: A cognitive presence case

In this paper, we present the results of an exploratory study that examined the problem of automating content analysis of student online discussion transcripts. We looked at the problem of coding discussion transcripts for the levels of cognitive presence, one of the three main constructs in the Community of Inquiry (CoI) model of distance education. Using Coh-Metrix and LIWC features, together with a set of custom features developed to capture discussion context, we developed a random forest classification system that achieved 70.3% classification accuracy and 0.63 Cohen's kappa, which is significantly higher than values reported in the previous studies. Besides improvement in classification accuracy, the developed system is also less sensitive to overfitting as it uses only 205 classification features, which is around 100 times less features than in similar systems based on bag-of-words features. We also provide an overview of the classification features most indicative of the different phases of cognitive presence that gives an additional insights into the nature of cognitive presence learning cycle. Overall, our results show great potential of the proposed approach, with an added benefit of providing further characterization of the cognitive presence coding scheme.

Designing against boredom in automated driving

I draw on four years of experience in mobility and transport research. I was part of a research project with Siemens, for which we identified global trends in urban mobility and explored future business opportunities through scenario planning methods. Some of the proposed solutions for personal and public transport included driverless vehicles. In collaboration with BMW Design I explored the potential of new materials for automotive user interfaces...

Infrared thermal imaging for automated detection of diabetic foot complications

Background Although thermal imaging can be a valuable technology in the prevention and management of diabetic foot disease, it is not yet widely used in clinical practice. Technological advancement in infrared imaging increases its application range. The aim was to explore the first steps in the applicability of high-resolution infrared thermal imaging for noninvasive automated detection of signs of diabetic foot disease. Methods The plantar foot surfaces of 15 diabetes patients were imaged with an infrared camera (resolution, 1.2 mm/pixel): 5 patients had no visible signs of foot complications, 5 patients had local complications (e.g., abundant callus or neuropathic ulcer), and 5 patients had difuse complications (e.g., Charcot foot, infected ulcer, or critical ischemia). Foot temperature was calculated as mean temperature across pixels for the whole foot and for specified regions of interest (ROIs). Results No diferences in mean temperature >1.5 °C between the ipsilateral and the contralateral foot were found in patients without complications. In patients with local complications, mean temperatures of the ipsilateral and the contralateral foot were similar, but temperature at the ROI was >2 °C higher compared with the corresponding region in the contralateral foot and to the mean of the whole ipsilateral foot. In patients with difuse complications, mean temperature diferences of >3 °C between ipsilateral and contralateral foot were found. Conclusions With an algorithm based on parameters that can be captured and analyzed with a high-resolution infrared camera and a computer, it is possible to detect signs of diabetic foot disease and to discriminate between no, local, or difuse diabetic foot complications. As such, an intelligent telemedicine monitoring system for noninvasive automated detection of signs of diabetic foot disease is one step closer. Future studies are essential to confirm and extend these promising early findings.

Computation of turbulent flow in an IFRF quarl burner

A computational model for isothermal axisymmetric turbulent flow in a quarl burner is set up using the CFD package FLUENT, and numerical solutions obtained from the model are compared with available experimental data. A standard k-e model and and two versions of the RNG k-e model are used to model the turbulence. One of the aims of the computational study is to investigate whether the RNG based k-e turbulence models are capable of yielding improved flow predictions compared with the standard k-e turbulence model. A difficulty is that the flow considered here features a confined vortex breakdown which can be highly sensitive to flow behaviour both upstream and downstream of the breakdown zone. Nevertheless, the relatively simple confining geometry allows us to undertake a systematic study so that both grid-independent and domain-independent results can be reported. The systematic study includes a detailed investigation of the effects of upstream and downstream conditions on the predictions, in addition to grid refinement and other tests to ensure that numerical error is not significant. Another important aim is to determine to what extent the turbulence model predictions can provide us with new insights into the physics of confined vortex breakdown flows. To this end, the computations are discussed in detail with reference to known vortex breakdown phenomena and existing theories. A major conclusion is that one of the RNG k-e models investigated here is able to correctly capture the complex forward flow region inside the recirculating breakdown zone. This apparently pathological result is in stark contrast to the findings of previous studies, most of which have concluded that either algebraic or differential Reynolds stress modelling is needed to correctly predict the observed flow features. Arguments are given as to why an isotropic eddy-viscosity turbulence model may well be able to capture the complex flow structure within the recirculating zone for this flow setup. With regard to the flow physics, a major finding is that the results obtained here are more consistent with the view that confined vortex breakdown is a type of axisymmetric boundary layer separation, rather than a manifestation of a subcritical flow state.

Visual detection of occluded crop: For automated harvesting

This paper presents a novel crop detection system applied to the challenging task of field sweet pepper (capsicum) detection. The field-grown sweet pepper crop presents several challenges for robotic systems such as the high degree of occlusion and the fact that the crop can have a similar colour to the background (green on green). To overcome these issues, we propose a two-stage system that performs per-pixel segmentation followed by region detection. The output of the segmentation is used to search for highly probable regions and declares these to be sweet pepper. We propose the novel use of the local binary pattern (LBP) to perform crop segmentation. This feature improves the accuracy of crop segmentation from an AUC of 0.10, for previously proposed features, to 0.56. Using the LBP feature as the basis for our two-stage algorithm, we are able to detect 69.2% of field grown sweet peppers in three sites. This is an impressive result given that the average detection accuracy of people viewing the same colour imagery is 66.8%.

Automated feature engineering for HTTP tunnel detection

Generating discriminative input features is a key requirement for achieving highly accurate classifiers. The process of generating features from raw data is known as feature engineering and it can take significant manual effort. In this paper we propose automated feature engineering to derive a suite of additional features from a given set of basic features with the aim of both improving classifier accuracy through discriminative features, and to assist data scientists through automation. Our implementation is specific to HTTP computer network traffic. To measure the effectiveness of our proposal, we compare the performance of a supervised machine learning classifier built with automated feature engineering versus one using human-guided features. The classifier addresses a problem in computer network security, namely the detection of HTTP tunnels. We use Bro to process network traffic into base features and then apply automated feature engineering to calculate a larger set of derived features. The derived features are calculated without favour to any base feature and include entropy, length and N-grams for all string features, and counts and averages over time for all numeric features. Feature selection is then used to find the most relevant subset of these features. Testing showed that both classifiers achieved a detection rate above 99.93% at a false positive rate below 0.01%. For our datasets, we conclude that automated feature engineering can provide the advantages of increasing classifier development speed and reducing development technical difficulties through the removal of manual feature engineering. These are achieved while also maintaining classification accuracy.

Automated inspection of aircraft parts using a modified ICP algorithm

The application of computer-aided inspection integrated with the coordinate measuring machine and laser scanners to inspect manufactured aircraft parts using robust registration of two-point datasets is a subject of active research in computational metrology. This paper presents a novel approach to automated inspection by matching shapes based on the modified iterative closest point (ICP) method to define a criterion for the acceptance or rejection of a part. This procedure improves upon existing methods by doing away with the following, viz., the need for constructing either a tessellated or smooth representation of the inspected part and requirements for an a priori knowledge of approximate registration and correspondence between the points representing the computer-aided design datasets and the part to be inspected. In addition, this procedure establishes a better measure for error between the two matched datasets. The use of localized region-based triangulation is proposed for tracking the error. The approach described improves the convergence of the ICP technique with a dramatic decrease in computational effort. Experimental results obtained by implementing this proposed approach using both synthetic and practical data show that the present method is efficient and robust. This method thereby validates the algorithm, and the examples demonstrate its potential to be used in engineering applications.

Automated Discovery of Structured Process Models: Discover Structured vs. Discover and Structure

This paper addresses the problem of discovering business process models from event logs. Existing approaches to this problem strike various tradeoffs between accuracy and understandability of the discovered models. With respect to the second criterion, empirical studies have shown that block-structured process models are generally more understandable and less error-prone than unstructured ones. Accordingly, several automated process discovery methods generate block-structured models by construction. These approaches however intertwine the concern of producing accurate models with that of ensuring their structuredness, sometimes sacrificing the former to ensure the latter. In this paper we propose an alternative approach that separates these two concerns. Instead of directly discovering a structured process model, we first apply a well-known heuristic technique that discovers more accurate but sometimes unstructured (and even unsound) process models, and then transform the resulting model into a structured one. An experimental evaluation shows that our “discover and structure” approach outperforms traditional “discover structured” approaches with respect to a range of accuracy and complexity measures.

Investigations of the scaling criteria for a mild combustion burner

In this paper, a new strategy for scaling burners based on "mild combustion" is evolved and adopted to scaling a burner from 3 to a 150 kW burner at a high heat release Late of 5 MW/m(3) Existing scaling methods (constant velocity, constant residence time, and Cole's procedure [Proc. Combust. Inst. 28 (2000) 1297]) are found to be inadequate for mild combustion burners. Constant velocity approach leads to reduced heat release rates at large sizes and constant residence time approach in unacceptable levels of pressure drop across the system. To achieve mild combustion at high heat release rates at all scales, a modified approach with high recirculation is adopted in the present studies. Major geometrical dimensions are scaled as D similar to Q(1/3) with an air injection velocity of similar to 100 m/s (Delta p similar to 600 mm water gauge). Using CFD support, the position of air injection holes is selected to enhance the recirculation rates. The precise role of secondary air is to increase the recirculation rates and burn LIP the residual CO in the downstream. Measurements of temperature and oxidizer concentrations inside 3 kW, 150 kW burner and a jet flame are used to distinguish the combustion process in these burners. The burner can be used for a wide range of fuels from LPG to producer gas as extremes. Up to 8 dB of noise level reduction is observed in comparison to the conventional combustion mode. Exhaust NO emissions below 26 and 3 ppm and temperatures 1710 and 1520 K were measured for LPG and producer gas when the burner is operated at stoichiometry. (c) 2004 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

An automated framework for understanding structural variations in the binding grooves of MHC class II molecules

Background: MHC/HLA class II molecules are important components of the immune system and play a critical role in processes such as phagocytosis. Understanding peptide recognition properties of the hundreds of MHC class II alleles is essential to appreciate determinants of antigenicity and ultimately to predict epitopes. While there are several methods for epitope prediction, each differing in their success rates, there are no reports so far in the literature to systematically characterize the binding sites at the structural level and infer recognition profiles from them. Results: Here we report a new approach to compare the binding sites of MHC class II molecules using their three dimensional structures. We use a specifically tuned version of our recent algorithm, PocketMatch. We show that our methodology is useful for classification of MHC class II molecules based on similarities or differences among their binding sites. A new module has been used to define binding sites in MHC molecules. Comparison of binding sites of 103 MHC molecules, both at the whole groove and individual sub-pocket levels has been carried out, and their clustering patterns analyzed. While clusters largely agree with serotypic classification, deviations from it and several new insights are obtained from our study. We also present how differences in sub-pockets of molecules associated with a pair of autoimmune diseases, narcolepsy and rheumatoid arthritis, were captured by PocketMatch(13). Conclusion: The systematic framework for understanding structuralvariations in MHC class II molecules enables large scale comparison of binding grooves and sub-pockets, which is likely to have direct implications towards predicting epitopes and understanding peptide binding preferences.

An automated framework for understanding structural variations in the binding grooves of MHC class II molecules

Background: MHC/HLA class II molecules are important components of the immune system and play a critical role in processes such as phagocytosis. Understanding peptide recognition properties of the hundreds of MHC class II alleles is essential to appreciate determinants of antigenicity and ultimately to predict epitopes. While there are several methods for epitope prediction, each differing in their success rates, there are no reports so far in the literature to systematically characterize the binding sites at the structural level and infer recognition profiles from them. Results: Here we report a new approach to compare the binding sites of MHC class II molecules using their three dimensional structures. We use a specifically tuned version of our recent algorithm, PocketMatch. We show that our methodology is useful for classification of MHC class II molecules based on similarities or differences among their binding sites. A new module has been used to define binding sites in MHC molecules. Comparison of binding sites of 103 MHC molecules, both at the whole groove and individual sub-pocket levels has been carried out, and their clustering patterns analyzed. While clusters largely agree with serotypic classification, deviations from it and several new insights are obtained from our study. We also present how differences in sub-pockets of molecules associated with a pair of autoimmune diseases, narcolepsy and rheumatoid arthritis, were captured by PocketMatch(13). Conclusion: The systematic framework for understanding structural variations in MHC class II molecules enables large scale comparison of binding grooves and sub-pockets, which is likely to have direct implications towards predicting epitopes and understanding peptide binding preferences.

Advanced transmission electron microscope triboprobe with automated closed-loop nanopositioning

Here the design and operation of a novel transmission electron microscope (TEM) triboprobe instrument with real-time vision control for advanced in situ electron microscopy is demonstrated. The NanoLAB triboprobe incorporates a new high stiffness coarse slider design for increased stability and positioning performance. This is linked with an advanced software control system which introduces both new and flexible in situ experimental functional testing modes, plus an automated vision control feedback system. This advancement in instrumentation design unlocks new possibilities of performing a range of new dynamical nanoscale materials tests, including novel friction and fatigue experiments inside the electron microscope.

Comparative study of filter-bank mean-energy distance for automated segmentation of speech signals

This paper describes a method of automated segmentation of speech assuming the signal is continuously time varying rather than the traditional short time stationary model. It has been shown that this representation gives comparable if not marginally better results than the other techniques for automated segmentation. A formulation of the 'Bach' (music semitonal) frequency scale filter-bank is proposed. A comparative study has been made of the performances using Mel, Bark and Bach scale filter banks considering this model. The preliminary results show up to 80 % matches within 20 ms of the manually segmented data, without any information of the content of the text and without any language dependence. 'Bach' filters are seen to marginally outperform the other filters.

Language independent automated segmentation of speech using Bach scale filter-banks

This correspondence describes a method for automated segmentation of speech. The method proposed in this paper uses a specially designed filter-bank called Bach filter-bank which makes use of 'music' related perception criteria. The speech signal is treated as continuously time varying signal as against a short time stationary model. A comparative study has been made of the performances using Mel, Bark and Bach scale filter banks. The preliminary results show up to 80 % matches within 20 ms of the manually segmented data, without any information of the content of the text and without any language dependence. The Bach filters are seen to marginally outperform the other filters.

Design and fabrication of an automated temperature programmed reaction system to evaluate 3-way catalysts Ce1-x-y(La/Y)(x)PtyO2-delta

A completely automated temperature-programmed reaction (TPR) system for carrying out gas-solid catalytic reactions under atmospheric flow conditions is fabricated to study CO and hydrocarbon oxidation, and NO reduction. The system consists of an all-stainless steel UHV system, quadrupole mass spectrometer SX200 (VG Scientific), a tubular furnace and micro-reactor, a temperature controller, a versatile gas handling system, and a data acquisition and analysis system. The performance of the system has been tested under standard experimental conditions for CO oxidation over well-characterized Ce1-x-y(La/Y)(y)O2-delta catalysts. Testing of 3-way catalysis with CO, NO and C2H2 to convert to CO2, N-2 and H2O is done with this catalyst which shows complete removal of pollutants below 325 degrees C. Fixed oxide-ion defects in Pt substituted Ce1-y(La/Y)(y)O2-y/2 show higher catalytic activity than Pt ion-substituted CeO2