Ken McGarry

Speech is a promising low-cost Alzheimer's disease marker, but in the DementiaBank Pitt Corpus, a recording-condition confound leaks into diagnostic labels. This confound is a systematic association between recording conditions and diagnosis, independent of speech features. Standard countermeasures include denoising and interviewer speech removal. We systematically evaluate both across nine denoisers and four speaker-isolation toolkits, using patient-level cross-validation. Because of the severe class imbalance, threshold-dependent metrics can conceal near-chance balanced accuracy; we then report prevalence-robust metrics. Final analysis found that no audio-quality metric predicts downstream discrimination, so quality cannot be used to select pre-processing methodologies. Recording-condition features alone predict diagnosis above chance (AUC 0.66 ± 0.02; patient-level permutation p < 0.001), and this persists after removing speech-quality features. Eight of nine denoisers fail to remove the confound; several amplify it; only Kalman filtering reduces it to chance. Denoising shifts downstream discrimination, but only three methods preserve the shift direction across both feature sets and classifiers, indicating that most apparent gains are specific to configuration. Speaker-isolation toolkits cannot remove the interviewer without discarding a substantial portion of the patient's speech. We found no remedy which would repair the confound. It is a corpus property and must be controlled at acquisition. This evaluation provides a reusable assessment tool for clinical speech pre-processing.

In this work we implement Bayesian surprise as a method to sift through sequences of discrete patterns and identify any unusual or interesting patterns that deviate from known sequences. Surprise is a biological trait inherent in humans and animals and is essential for many creative acts and efforts of discovery. Numerous technical domains are comprised of discrete elements in sequences such as e‐commerce transactions, genome data searching, online financial transactions of many types, criminal cyber‐attacks and life‐course data from sociology. In addition to the complexity and computational burden of this type of problem is the issue of their rarity. Many anomalies are infrequent and may defy categorisation; therefore, they are not suited to classification solutions. We test our methods on four discrete datasets (Hospital Sepsis patients, Chess Moves, the Wisconsin Card Sorting Task and BioFamilies) consisting of discrete sequences. Probabilistic Suffix Trees are trained on this data which maintain each discrete symbol's location and position in a given sequence. The trained models are exposed to “new” data where any deviations from learned patterns either in location on the sequence or frequency of occurrence will denote patterns that are unusual compared with the original training data. To assist in the identification of new patterns and to avoid confusing old patterns as new or novel we use Bayesian surprise to detect the discrepancies between what we are expecting and actual results. We can assign the degree of surprise or unexpectedness to any new pattern and provide an indication of why certain patterns are deemed novel or surprising and why others are not.

In this work we combine sentiment analysis with graph theory to analyze user posts, likes/dislikes on a variety of social media to provide recommendations for YouTube videos. We focus on the topic of climate change/global warming, which has caused much alarm and controversy over recent years. Our intention is to recommend informative YouTube videos to those seeking a balanced viewpoint of this area and the key arguments/issues. To this end we analyze Twitter data; Reddit comments and posts; user comments, view statistics and likes/dislikes of YouTube videos. The combination of sentiment analysis with raw statistics and linking users with their posts gives deeper insights into their needs and quest for quality information. Sentiment analysis provides the insights into user likes and dislikes, graph theory provides the linkage patterns and relationships between users, posts, and sentiment.

The research aimed to develop a statistical analysis of the antifouling properties of nanocomposite membrane using response surface methodology (RSM) for methyl orange (MO) dye wastewater treatment. Reduced graphene oxide (rGO) and multi-walled carbon nanotubes (MWCNTs) were blended directly into polyvinylidene difluoride (PVDF) membrane. 13 experiment runs were conducted using central composite design (CCD). Nanomaterials concentration and weight ratio of rGO:MWCNTs were manipulated, while the response was normalized flux. Next, the membranes were characterized by pore size, porosity, surface charge, and hydrophilicity. In general, the increase in membrane pore size was mainly due to the addition of nanomaterials. Lastly, the membrane performance was evaluated using dead-end filtration. The best antifouling was achieved by membrane containing 0.023 wt% nanomaterial concentration and 85.36 wt% weight ratio with the highest normalized flux of 0.9132. The antifouling properties of nanocomposite membrane were then modeled statistically using analysis of variance (ANOVA) and regression analysis to identify the data significance. Lastly, a quadratic model equation was rendered and it is found that normalized flux was significantly affected by nanomaterials concentration. The model was proven to be fitted with moderate accuracy based on the F-value (4.09), p-values (0.0468 2 value (0.7449), and standard deviation (0.0946).

BackgroundIn this work we employ Bayesian surprise to detect interesting/anomalous patterns from discrete sequence data.Many domains consist of discrete sequential time-series such as DNA analysis, online transactions, web click-stream navigation, cyber-attacks, financial transactions and especially sociology life-course data. The difficulty is that each data set has its own unique characteristics and many anomalies defy categorization. Since anomalies are by nature infrequent and elusive, we often do not have enough data for a supervised approach. However, novelty and surprise play a fundamental role in human and animal behavior for survival, attention and adaptation.MethodsWe use three sequence datasets (Swiss Health, Sepsis, and BioFamilies) which are are composed of simpler motifs which are used to build Probabilistic Suffix Trees (PST) which can capture complex relationships based on motif location and frequency of occurrence. New data that deviates from established motifs either in location of appearance, frequency of appearance, or motif composition may represent recurring patterns that may be different in some way. Bayesian surprise is the result of mismatches between our expectations and actual results, hence the degree of surprise or anomalousness attached to a pattern will vary with respect to these differences.ResultsEach data set is assessed by Bayesian Surprise and several other criteria, providing indications of why certain patterns are interesting and why others are not.ConclusionsBayesian surprise can detect data with other properties that would be missed by information theoretic measures such as Shannon surprise and entropy for example.