Nick Cercone's research while affiliated with York University and other places

Demand forecasting is one of the important inputs for a successful restaurant yield and revenue management system. Sales forecasting is crucial for an independent restaurant and for restaurant chains as well. In this chapter a comprehensive literature review and classification of restaurant sales and consumer demand techniques are presented. Sales prediction is very complex due to the impact of internal and external environment. However, a reliable sales forecasting methodology can improve the quality of business strategy. A range of methodologies and models for forecasting are given in the literature. These techniques are categorized here into seven categories, also including hybrid models. The methodology for different kinds of analytical methods is briefly described, the advantages and drawbacks are discussed, and relevant set of papers is selected. Conclusions and comments are also made on future research directions.

MapReduce is an important programming model for processing in distributed environments. Compared to other distributed programming models, MapReduce reduces communication overheads between computers and improves fault tolerance. However, the MapReduce model does not allow for automatic synchronization between jobs. A large number of data analytics algorithms use a recursive divide-and-conquer approach, which inherently allows for parallelism at each level of recursion. However, it is often difficult to parallelize such algorithms using the traditional MapReduce model if the process requires synchronization. In this paper we introduce Parent-Child MapReduce, a version of the MapReduce programming model that allows for MapReduce tasks to be created dynamically and synchronized in a hierarchical parent-child fashion. Using the Parallel FP-Growth (PFP) algorithm for mining frequent patterns as a reference, we show that Parent-Child MapReduce can be used to parallelize recursive divide-and-conquer algorithms using the MapReduce model and that this can lead to significant speed ups in the computational speed of such algorithms. Our evaluation shows that we can achieve 68% (or 3 times) performance gain when used with PFP.

Demand forecasting is one of the important inputs for a successful restaurant yield and revenue management system. Sales forecasting is crucial for an independent restaurant and for restaurant chains as well. In the paper a comprehensive literature review and classification of restaurant sales and consumer demand techniques are presented. A range of methodologies and models for forecasting are given in the literature. These techniques are categorized here into seven categories, also included hybrid models. The methodology for different kind of analytical methods is briefly described, the advantages and drawbacks are discussed, and relevant set of papers is selected. Conclusions and comments are also made on future research directions.

In the Canadian Province of Ontario, electricity consumers pay a surcharge for electricity called the Global Adjustment (GA). For large consumers, having the ability to predict the top 5 daily energy demand hours of the year, called 5 Coincident Peaks (5CPs), can save millions of dollars in GA costs, and help decrease peak energy usage. This paper presents a Naive Bayesian classification model for predicting the 5CPs. The model classifies hourly energy demand as being a 5CP hour or not. The model was tested using hourly energy demand for the province of Ontario over a 21 year period (1995–2015). Classifying a day as a 5CP hour containing day yielded a mean precision and recall of 0.49 (0.18) and 0.88 (0.23) (Standard deviation is in brackets), respectively. Targeting the 5CP hours to within three candidate hours of potential 5CP containing days yielded a mean precision and recall of 0.47 (0.19) and 0.83 (0.22), respectively.

Within e-learning environment and research, adaptive learning systems (ALS) can provide an adaptive learning environment that intelligently adapts to a user's needs by presenting suitable information, instructional materials, feedback, and recommendations based on the user's unique individual characteristics and situation. This chapter discusses three main contributions of data mining to adaptive learning systems (ALS), including prediction, classification, and recommendation. One of the most special features of ALS is the user model that represents essential personalized information about each user. The ALS model consists of the storage module, the user model module, the reasoning module, and the management module. The reasoning module's main functions are to support student learning activities. This module is responsible for filtering knowledge points (KP's) filtering, searching and organizing learning resources, generating a learning program, and updating the user model and the knowledge base.

Background: Complications from type 2 diabetes mellitus can be prevented when patients perform health behaviors such as vigorous exercise and glucose-regulated diet. The use of smartphones for tracking such behaviors has demonstrated success in type 2 diabetes management while generating repositories of analyzable digital data, which, when better understood, may help improve care. Data mining methods were used in this study to better understand self-monitoring patterns using smartphone tracking software. Objective: Associations were evaluated between the smartphone monitoring of health behaviors and HbA1c reductions in a patient subsample with type 2 diabetes who demonstrated clinically significant benefits after participation in a randomized controlled trial. Methods: A priori association-rule algorithms, implemented in the C language, were applied to app-discretized use data involving three primary health behavior trackers (exercise, diet, and glucose monitoring) from 29 participants who achieved clinically significant HbA1c reductions. Use was evaluated in relation to improved HbA1c outcomes. Results: Analyses indicated that nearly a third (9/29, 31%) of participants used a single tracker, half (14/29, 48%) used two primary trackers, and the remainder (6/29, 21%) of the participants used three primary trackers. Decreases in HbA1c were observed across all groups (0.97-1.95%), but clinically significant reductions were more likely with use of one or two trackers rather than use of three trackers (OR 0.18, P=.04). Conclusions: Data mining techniques can reveal relevant coherent behavior patterns useful in guiding future intervention structure. It appears that focusing on using one or two trackers, in a symbolic function, was more effective (in this sample) than regular use of all three trackers.

The image feature used for classification is a crucial part of a character recognition system. To achieve a high accuracy of offline handwriting recognition, the feature should capture the essence of differences including the differences between different characters and the differences between different drawings of the same character. In this paper, we present a novel image feature called direction histogram (DH) and a feature extraction algorithm called bag of histogram (BoH). Unlike the traditional pre-defined feature, DH was designed based on the nature of language and the variation of writing styles. DH is, therefore, a global feature that represents pixel density in all directions around each center. BoH was introduced as it tolerates to thickness and curve variation and ignores the curve connectivity (if any). Fifty-two datasets, each containing 30 drawings of 80 Thai characters, are used for training our neural network, and the original, thick, and distorted handwriting datasets are used for testing. The recognition system with our proposed DH and BoH feature extraction algorithm yielded higher recognition accuracy compared to the convolutional neural network.