B. Sumudu U. Mendis

A Hybrid Fuzzy Approach for Human Eye Gaze Pattern Recognition

Face perception and text reading are two of the most developed visual perceptual skills in humans. Understanding which features in the respective visual patterns make them differ from each other is very important for us to investigate the correlation between humans visual behavior and cognitive processes. We introduce our fuzzy signatures with a Levenberg-Marquardt optimization method based hybrid approach for recognizing the different eye gaze patterns when a human is viewing faces or text documents. Our experimental results show the effectiveness of using this method for the real world case. A further comparison with Support Vector Machines (SVM) also demonstrates that by defining the classification process in a similar way to SVM, our hybrid approach is able to provide a comparable performance but with a more interpretable form of the learned structure.

Eye gaze assistance for a game-like interactive task

Human beings communicate in abbreviated ways dependent on prior interactions and shared knowledge. Furthermore, humans share information about intentions and future actions using eye gaze. Among primates, humans are unique in the whiteness of the sclera and amount of sclera shown, essential for communication via interpretation of eye gaze. This paper extends our previous work in a game-like interactive task by the use of computerised recognition of eye gaze and fuzzy signature-based interpretation of possible intentions. This extends our notion of robot instinctive behaviour to intentional behaviour.We show a good improvement of speed of response in a simple use of eye gaze information.We also show a significant and more sophisticated use of the eye gaze information, which eliminates the need for control actions on the users part. We also make a suggestion as to returning visibility of control to the user in these cases.

Predicting reading comprehension scores from eye movements using artificial neural networks and fuzzy output error

Predicting reading comprehension from eye gaze data is a difficult task. We investigate the use of artificial neural networks(ANNs) to predict reading comprehension scores from eye gaze collected from participants who read and completed an onlinetutorial in our lab. Problems such as large feature sets and small highly imbalanced data sets compound to make this task evenmore complex. We propose using fuzzy output error (FOE) as an alternative performance function to mean square error (MSE)for training feed-forward neural networks to overcome these problems. We show that the use of FOE as the performance functionfor training ANNs provides significantly better classification of eye movements to reading comprehension scores. ANNs withthree hidden layers of neurons gave the best classification results especially when FOE is used as the performance functionfor training. In these cases we found up to 50% reduction in misclassification rates compared to using MSE. We found thatANNs give optimal classification results in comparison to other classification techniques. When FOE is used as the performancefunction for training the ANNs the misclassification rates are halved compared to the other techniques. Cluster analysis wasperformed on one of the more complex data sets. Interesting reading behaviour properties were found within the data set.The intended use of this research is in the design of adaptive online learning environments that use eye gaze to predict usercomprehension from reading behavior.

Fuzzy word similarity: A semantic approach using WordNet

In this paper we present a hybrid measure of semantic word similarity using fuzzy inference system which combines both the corpus based distance measures as well as gloss overlap to get the final similarity between two words. We use WordNet as a lexical dictionary to get semantic information about words. We show that this new measure reasonably correlates to human judgments and the average performance is boosted by using triangular membership function in the output.

Hierarchical document signature: A specialized application of fuzzy signature for document computing

We develop document computing procedures for the analysis of discourse structures within a document, represented by hierarchical document signatures. A signature is a string of data characterizing a certain case (e.g. characteristics of a sentence in case of a document). The place of the individual data is fixed within the string, it holds a local value semantics. Fuzzy granulation is a semantic background technique for all kinds of information which originates from human estimation or recorded by human valuation of numerical data. For analysis of such data the development of special procedures is suggested, different from the usual statistical methods. We used a form of fuzzy signature, called hierarchical document signature to modularize an unstructured document in a hierarchical manner, from Document level to sentence level, sentence level to attribute level and then to word level. We used occurrence of words as the information of the lowest module to find the similarity among the next higher module by aggregating the signature values giving sentence pair coherence.

Fuzzy Output Error as the Performance Function for Training Artificial Neural Networks to Predict Reading Comprehension from Eye Gaze

Imbalanced data sets are common in real life and can have a negative effect on classifier performance. We propose using fuzzy output error (FOE) as an alternative performance function to mean square error (MSE) for training feed forward neural networks to overcome this problem. The imbalanced data sets we use are eye gaze data recorded from reading and answering a tutorial and quiz. The goal is to predict the quiz scores for each tutorial page. We show that the use of FOE as the performance function for training neural networks provides significantly better classification of eye movements to reading comprehension scores. A neural network with three hidden layers of neurons gave the best classification results especially when FOE was used as the performance function for training. In these cases, upwards of a 19% reduction in misclassification was achieved compared to using MSE as the performance function.

Evolutionary algorithms using cluster patterns for timetabling

The examination timetabling problem ETP is a NP complete, combinatorial optimization problem. Intuitively, use of properties such as patterns or clusters in the data suggests possible improvements in the performance and quality of timetabling. This paper investigates whether the use of a genetic algorithm GA informed by patterns extracted from student timetable data to solve ETPs can produce better quality solutions. The data patterns were captured in clusters, which then were used to generate the initial population and evaluate fitness of individuals. The proposed techniques were compared with a traditional GA and popular techniques on widely used benchmark problems, and a local data set, the Australian National University ANU ETP, which was the motivating problem for this work. A formal definition of the ANU ETP is also proposed. Results show techniques using cluster patterns produced better results than the traditional GA with statistical significance of p < 0.01, showing strong evidence. Our techniques either clearly outperformed or performed well compared to the best known techniques in the literature and produced a better timetable than the manually constructed timetable used by ANU, both in terms of quality and execution time. In this work, we also propose clear criteria for specifying the top results in this area.

Improving hierarchical document signature performance by classifier combination

We present a classifier-combination experimental framework for part-of-speech (POS) tagging in which four different POS taggers are combined in order to get a better result for sentence similarity using Hierarchical Document Signature (HDS). It is important to abstract information available to form humanly accessible structures. The way people think and talk is hierarchical with limited information presented in any one sentence, and that information is always linked together to further information. As such, HDS is a significant way to represent sentences when finding their similarity. POS tagging plays an important role in HDS. But POS taggers available are not perfect in tagging words in a sentence and tend to tag words improperly if they are either not properly cased or do not match the corpus dataset by which these taggers are trained. Thus, different weighted voting strategies are used to overcome some of these drawbacks of these existing taggers. Comparisons between individual taggers and combined taggers under different voting strategies are made. Their results show that the combined taggers provide better results than the individual ones.

Fuzzy methods and eye gaze for cooperative robot communication

We introduce our fuzzy signature and pattern matching with possibility calculation based approach for modelling the communication between human-control robot and a pair of assistant robots for completing cooperative tasks in a simulated environment. We show that a sophisticated extension using computerised recognition of eye gaze with fuzzy modelling based interpretation of possible intentions effectively eliminates the need of any physical control from the humans side. The experiment results show a good improvement of time saved by the use of our eye gaze intention in the context.