Philip Weber

A Framework for the Analysis of Process Mining Algorithms

There are many process mining algorithms and representations, making it difficult to choose which algorithm to use or compare results. Process mining is essentially a machine learning task, but little work has been done on systematically analyzing algorithms to understand their fundamental properties, such as how much data are needed for confidence in mining. We propose a framework for analyzing process mining algorithms. Processes are viewed as distributions over traces of activities and mining algorithms as learning these distributions. We use probabilistic automata as a unifying representation to which other representation languages can be converted. We present an analysis of the Alpha algorithm under this framework and experimental results, which show that from the substructures in a model and behavior of the algorithm, the amount of data needed for mining can be predicted. This allows efficient use of data and quantification of the confidence which can be placed in the results.

A framework for comparing process mining algorithms

There are many process mining algorithms with different theoretical foundations and aims, raising the question of how to choose the best for a particular situation. A framework is proposed for objectively comparing algorithms for process discovery against a known ground truth, with an implementation using existing tools. Results from an experimental evaluation of five algorithms against basic process structures confirm the validity of the approach. In general, numbers of traces for mining are predictable from the structure and probabilities in the model, but there are some algorithm-specific differences.

Trajectory analysis of speech using continuous state hidden Markov Models

Many current speech models used in recognition involve thousands of parameters, whereas the mechanisms of speech production are conceptually very simple. We present and evaluate a new continuous state probabilistic model (CS-HMM) for recovering dwell-transition and phoneme sequences from dynamic speech production features. We show that with very few parameters, these features can be tracked, and phoneme sequences recovered, with promising accuracy.

Assessing the extent of drug interactions among patients with multimorbidity in primary and secondary care in the West Midlands (UK): a study protocol for the Mixed Methods Multimorbidity Study (MiMMS)

Introduction The numbers of patients with three or more chronic conditions (multimorbidity) are increasing, and will rise to 2.9 million by 2018 in the UK alone. Currently in the UK, conditions are mainly managed using over 250 sets of single-condition guidance, which has the potential to generate conflicting recommendations for lifestyle and concurrent medication for individual patients with more than one condition. To address some of these issues, we are developing a new computer-based tool to help manage these patients more effectively. For this tool to be applicable and relevant to current practice, we must first better understand how existing patients with multimorbidity are being managed, particularly relating to concerns over prescribing and potential polypharmacy. Methods and analysis Up to four secondary care centres, two community pharmacies and between four and eight primary care centres in the West Midlands will be recruited. Interviewees will be purposively sampled from these sites, up to a maximum of 30. In this mixed methods study, we will perform a dual framework analysis on the qualitative data; the first analysis will use the Theoretical Domains Framework to assess barriers and enablers for healthcare professionals around the management of multimorbid patients; the second analysis will use Normalisation Process Theory to understand how interventions are currently being successfully implemented in both settings. We will also extract quantitative anonymised patient data from primary care to determine the extent of polypharmacy currently present for patients with multimorbidity in the West Midlands. Discussion We aim to combine these data so that we can build a useful, fully implementable tool which addresses the barriers most amenable to change within both primary and secondary care contexts. Ethics and dissemination Favourable ethical approval has been granted by The University of Birmingham Research Ethics Committee (ERN_16–0074) on 17 May 2016. Our work will be disseminated through peer-reviewed literature, trade journals and conferences. We will also use the dedicated web page hosted by the University to serve as a central point of contact and as a repository of our findings. We aim to produce a minimum of three articles from this work to contribute to the international scientific literature. Protocol registration number NIHR Clinical Research Network Portfolio Registration CPMS ID 30613.

Progress on phoneme recognition with a continuous-state HMM

Recent advances in automatic speech recognition have used large corpora and powerful computational resources to train complex statistical models from high-dimensional features, to attempt to capture all the variability found in natural speech. Such models are difficult to interpret and may be fragile, and contradict or ignore knowledge of human speech production and perception. We report progress towards phoneme recognition using a model of speech which employs very few parameters and which is more faithful to the dynamics and model of human speech production. Using features generated from a neural network bottleneck layer, we obtain recognition accuracy on TIMIT which compares favourably with traditional models of similar power. We discuss the implications of these results for recognition using natural features such as vocal tract resonances and spectral energies.

Interpretation of Low Dimensional Neural Network Bottleneck Features in Terms of Human Perception and Production.

Low-dimensional ‘bottleneck’ features extracted from neural networks have been shown to give phoneme recognition accuracy similar to that obtained with higher-dimensional MFCCs, using GMM-HMM models. Such features have also been shown to preserve well the assumptions of speech trajectory dynamics made by dynamic models of speech such as ContinuousState HMMs. However, little is understood about how networks derive these features and how and whether they can be interpreted in terms of human speech perception and production. We analyse three-dimensional bottleneck features. We show that for vowels, their spatial representation is very close to the familiar F1:F2 vowel quadrilateral. For other classes of phonemes the features can similarly be related to phonetic and acoustic spatial representations presented in the literature. This suggests that these networks derive representations specific to particular phonetic categories, with properties similar to those used by human perception. The representation of the full set of phonemes in the bottleneck space is consistent with a hypothesized comprehensive model of speech perception and also with models of speech perception such as prototype theory.

Automated Conflict Detection Between Medical Care Pathways

Clinical guidelines specify sequences of steps (care pathways) to treat patients with single conditions. Increasingly, many patients exhibit “multimorbidity,” several chronic conditions needing concurrent treatment. However, applying multiple guidelines in parallel can lead to conflicts, eg, between prescribed drugs, lifestyle intervention recommendations, or treatment schedules. In computer science, process languages used to design and reason about software development and business process management are similar to clinical pathways. Using formal model transformation, composition and analysis methods, models can be combined and conflicts detected and resolved. We propose BPMN+V, a data‐driven formal model for clinical care pathways, as an extension of Business Process Model and Notation. We describe a method for conflict detection using a transformation of BPMN+V to Coloured Petri Nets and a state‐space method for detection of conflict in composed models. We present results from a case study, showing that common conflicts are successfully detected, and propose extension to a complete framework for efficiently recommending resolutions to medical conflicts in composed care pathway models.