Shehroz S. Khan

Cluster center initialization algorithm for K -means clustering

Performance of iterative clustering algorithms which converges to numerous local minima depend highly on initial cluster centers. Generally initial cluster centers are selected randomly. In this paper we propose an algorithm to compute initial cluster centers for K-means clustering. This algorithm is based on two observations that some of the patterns are very similar to each other and that is why they have same cluster membership irrespective to the choice of initial cluster centers. Also, an individual attribute may provide some information about initial cluster center. The initial cluster centers computed using this methodology are found to be very close to the desired cluster centers, for iterative clustering algorithms. This procedure is applicable to clustering algorithms for continuous data. We demonstrate the application of proposed algorithm to K-means clustering algorithm. The experimental results show improved and consistent solutions using the proposed algorithm.

A survey of recent trends in one class classification

The One Class Classification (OCC) problem is different from the conventional binary/multi-class classification problem in the sense that in OCC, the negative class is either not present or not properly sampled. The problem of classifying positive (or target) cases in the absence of appropriately-characterized negative cases (or outliers) has gained increasing attention in recent years. Researchers have addressed the task of OCC by using different methodologies in a variety of application domains. In this paper we formulate a taxonomy with three main categories based on the way OCC has been envisaged, implemented and applied by various researchers in different application domains. We also present a survey of current state-of-the-art OCC algorithms, their importance, applications and limitations.

One-class classification: taxonomy of study and review of techniques

One-class classification (OCC) algorithms aim to build classification models when the negative class is either absent, poorly sampled or not well defined. This unique situation constrains the learning of efficient classifiers by defining class boundary just with the knowledge of positive class. The OCC problem has been considered and applied under many research themes, such as outlier/ novelty detection and concept learning. In this paper, we present a unified view of the general problem of OCC by presenting a taxonomy of study for OCC problems, which is based on the availability of training data, algorithms used and the application domains applied. We further delve into each of the categories of the proposed taxonomy and present a comprehensive literature review of the OCC algorithms, techniques and methodologies with a focus on their significance, limitations and applications. We conclude our paper by discussing some open research problems in the field of OCC and present our vision for future research.

Cluster center initialization algorithm for K-modes clustering

Partitional clustering of categorical data is normally performed by using K-modes clustering algorithm, which works well for large datasets. Even though the design and implementation of K-modes algorithm is simple and efficient, it has the pitfall of randomly choosing the initial cluster centers for invoking every new execution that may lead to non-repeatable clustering results. This paper addresses the randomized center initialization problem of K-modes algorithm by proposing a cluster center initialization algorithm. The proposed algorithm performs multiple clustering of the data based on attribute values in different attributes and yields deterministic modes that are to be used as initial cluster centers. In the paper, we propose a new method for selecting the most relevant attributes, namely Prominent attributes, compare it with another existing method to find Significant attributes for unsupervised learning, and perform multiple clustering of data to find initial cluster centers. The proposed algorithm ensures fixed initial cluster centers and thus repeatable clustering results. The worst-case time complexity of the proposed algorithm is log-linear to the number of data objects. We evaluate the proposed algorithm on several categorical datasets and compared it against random initialization and two other initialization methods, and show that the proposed method performs better in terms of accuracy and time complexity. The initial cluster centers computed by the proposed approach are close to the actual cluster centers of the different data we tested, which leads to faster convergence of K-modes clustering algorithm in conjunction to better clustering results.

Evaluating visual aesthetics in photographic portraiture

We propose and demonstrate a strategy to quantify aesthetic quality in photographs. Our approach is to develop a small set of classification features by tuning general compositional principles to a targeted image domain where saliency can be better understood. We demonstrate this strategy with photographic portraits of individuals, but it can be extended to other domains. Our technique leverages a refined method of using templates as spatial composition feature look-up tables. Compared to the traditional approach using a large set of global and local features extracted with little salient knowledge, classifiers using features extracted with our approach are better predictors of human aesthetic judgments.

Towards the detection of unusual temporal events during activities using HMMs

Most of the systems for recognition of activities aim to identify a set of normal human activities. Data is either recorded by computer vision or sensor based networks. These systems may not work properly if an unusual event or abnormal activity occurs, especially ones that have not been encountered in the past. By definition, unusual events are mostly rare and unexpected, and therefore very little or no data may be available for training. In this paper, we focus on the challenging problem of detecting unusual temporal events in a sensor network and present three Hidden Markov Models (HMM) based approaches to tackle this problem. The first approach models each normal activity separately as an HMM and the second approach models all the normal activities together as one common HMM. If the likelihood is lower than a threshold, an unusual event is identified. The third approach models all normal activities together in one HMM and approximates an HMM for the the unusual events. All the methods train HMM models on data of the usual events and do not require training data from the unusual events. We perform our experiments on a Locomotion Analysis dataset that contains gyroscope, force sensor, and accelerometer readings. To test the performance of our approaches, we generate five types of unusual events that represent random activity, extremely unusual events, unusual events similar to specific normal activities, no or little motion and normal activity followed by no or little motion. Our experiments suggest that for a moderately sized time frame window, these approaches can identify all the five types of unusual events with high confidence.

Review of fall detection techniques: A data availability perspective

A fall is an abnormal activity that occurs rarely; however, missing to identify falls can have serious health and safety implications on an individual. Due to the rarity of occurrence of falls, there may be insufficient or no training data available for them. Therefore, standard supervised machine learning methods may not be directly applied to handle this problem. In this paper, we present a taxonomy for the study of fall detection from the perspective of availability of fall data. The proposed taxonomy is independent of the type of sensors used and specific feature extraction/selection methods. The taxonomy identifies different categories of classification methods for the study of fall detection based on the availability of their data during training the classifiers. Then, we present a comprehensive literature review within those categories and identify the approach of treating a fall as an abnormal activity to be a plausible research direction. We conclude our paper by discussing several open research problems in the field and pointers for future research.

Relational approach to knowledge engineering for POMDP-based assistance systems as a translation, of a psychological model

Abstract Assistive systems for persons with cognitive disabilities (e.g., dementia) are difficult to build due to the wide range of different approaches people can take to accomplishing the same task, and the significant uncertainties that arise from both the unpredictability of client’s behaviours and from noise in sensor readings. Partially observable Markov decision process (POMDP) models have been used successfully as the reasoning engine behind such assistive systems for small multi-step tasks such as hand washing. POMDP models are a powerful, yet flexible framework for modelling assistance that can deal with uncertainty and utility. Unfortunately, POMDPs usually require a very labour intensive, manual procedure for their definition and construction. Our previous work has described a knowledge driven method for automatically generating POMDP activity recognition and context sensitive prompting systems for complex tasks. We call the resulting POMDP a SNAP (SyNdetic Assistance Process). The spreadsheet-like result of the analysis does not correspond to the POMDP model directly and the translation to a formal POMDP representation is required. To date, this translation had to be performed manually by a trained POMDP expert. In this paper, we formalise and automate this translation process using a probabilistic relational model (PRM) encoded in a relational database. The database encodes the relational skeleton of the PRM, and includes the goals, action preconditions, environment states, cognitive model, client and system actions (i.e., the outcome of the SNAP analysis), as well as relevant sensor models. The database is easy to approach for someone who is not an expert in POMDPs, allowing them to fill in the necessary details of a task using a simple and intuitive procedure. The database, when filled, implicitly defines a ground instance of the relational skeleton, which we extract using an automated procedure, thus generating a POMDP model of the assistance task. A strength of the database is that it allows constraints to be specified, such that we can verify the POMDP model is, indeed, valid for the task given the analysis. We demonstrate the method by eliciting three assistance tasks from non-experts: handwashing, and toothbrushing for elderly persons with dementia, and on a factory assembly task for persons with a cognitive disability. We validate the resulting POMDP models using case-based simulations to show that they are reasonable for the domains. We also show a complete case study of a designer specifying one database, including an evaluation in a real-life experiment with a human actor.

Detecting falls with X-Factor Hidden Markov Models

Graphical abstractDisplay Omitted HighlightsProposed new X-Factor Hidden Markov Models to identify falls in the absence of their training data.Proposed a novel cross-validation method to optimize parameters in the absence of fall data.Experimentally showed that performance of supervised classifiers deteriorate with very limited training fall data. Identification of falls while performing normal activities of daily living (ADL) is important to ensure personal safety and well-being. However, falling is a short term activity that occurs rarely and infrequently. This poses a challenge for traditional supervised classification algorithms, because there may be very little training data for falls (or none at all) to build generalizable models for falls. This paper proposes an approach for the identification of falls using a wearable device in the absence of training data for falls but with plentiful data for normal ADL. We propose three X-Factor Hidden Markov Model (XHMMs) approaches. The XHMMs have inflated output covariances (observation models). To estimate the inflated covariances, we propose a novel cross validation method to remove outliers from the normal ADL that serves as proxies for the unseen falls and allow learning the XHMMs using only normal activities. We tested the proposed XHMM approaches on two activity recognition datasets and show high detection rates for falls in the absence of fall-specific training data. We show that the traditional method of choosing threshold based on maximum of negative of log-likelihood to identify unseen falls is ill-posed for this problem. We also show that supervised classification methods perform poorly when very limited fall data is available during the training phase.

X-Factor HMMs for Detecting Falls in the Absence of Fall-Specific Training Data

Detection of falls is very important from a health and safety perspective. However, falls occur rarely and infrequently, which leads to either limited or no training data and thus can severely impair the performance of supervised activity recognition algorithms. In this paper, we address the problem of identification of falls in the absence of training data for falls, but with abundant training data for normal activities. We propose two ‘X-Factor’ Hidden Markov Model (XHMMs) approaches that are like normal HMMs, but have “inflated” output covariances (observation models), which can be estimated using cross-validation on the set of ‘outliers’ in the normal data that serve as proxies for the (unseen) fall data. This allows the XHMMs to be learned from only normal activity data. We tested the proposed XHMM approaches on two real activity recognition datasets that show high detection rates for falls in the absence of training data.