We introduce an extended representation of time series that allows fast, accurate classification and clustering in addition to the ability to explore time series data in a relevance feedback framework. The representation consists of piece-wise linear segments to represent shape and a weight vector that contains the relative importance of each individual linear segment. In the classification context, the weights are learned automatically as part of the training cycle. In the relevance feedback context, the weights are determined by an interactive and iterative process in which users rate various choices presented to them. Our representation allows a user to define a variety of similarity measures that can be tailored to specific domains. We demonstrate our approach on space telemetry, medical and synthetic data.

/pdf/an-enhanced-representation-of-time-series-which-allows-fast-1rsx7j7w9n.pdf

An enhanced representation of time series which allows fast and accurate classification, clustering and relevance feedback

This paper outlines an investigation of the readability of the chairman's narratives, as determined by the Flesch score, of a limited company's corporate reports from 1952–1985. It supports earlier studies which demonstrated that corporate reports were difficult to read, and had become more difficult to read over time. Turnover and readability were significantly negatively correlated. There also seemed to be a decrease in readability associated with the firm becoming a public limited company.

A Longitudinal Study of the Readability of the Chairman's Narratives in the Corporate Reports of a UK Company

Purpose




The purpose of this paper is to advance knowledge of the transformative potential of big data on city-based transport models. The central question guiding this paper is: how could big data transform smart city transport operations? In answering this question the authors present initial results from a Markov study. However the authors also suggest caution in the transformation potential of big data and highlight the risks of city and organizational adoption. A theoretical framework is presented together with an associated scenario which guides the development of a Markov model.




Design/methodology/approach




A model with several scenarios is developed to explore a theoretical framework focussed on matching the transport demands (of people and freight mobility) with city transport service provision using big data. This model was designed to illustrate how sharing transport load (and capacity) in a smart city can improve efficiencies in meeting demand for city services.




Findings




This modelling study is an initial preliminary stage of the investigation in how big data could be used to redefine and enable new operational models. The study provides new understanding about load sharing and optimization in a smart city context. Basically the authors demonstrate how big data could be used to improve transport efficiency and lower externalities in a smart city. Further how improvement could take place by having a car free city environment, autonomous vehicles and shared resource capacity among providers.




Research limitations/implications




The research relied on a Markov model and the numerical solution of its steady state probabilities vector to illustrate the transformation of transport operations management (OM) in the future city context. More in depth analysis and more discrete modelling are clearly needed to assist in the implementation of big data initiatives and facilitate new innovations in OM. The work complements and extends that of Setia and Patel (2013), who theoretically link together information system design to operation absorptive capacity capabilities.




Practical implications




The study implies that transport operations would actually need to be re-organized so as to deal with lowering CO2 footprint. The logistic aspects could be seen as a move from individual firms optimizing their own transportation supply to a shared collaborative load and resourced system. Such ideas are radical changes driven by, or leading to more decentralized rather than having centralized transport solutions (Caplice, 2013).




Social implications




The growth of cities and urban areas in the twenty-first century has put more pressure on resources and conditions of urban life. This paper is an initial first step in building theory, knowledge and critical understanding of the social implications being posed by the growth in cities and the role that big data and smart cities could play in developing a resilient and sustainable transport city system.




Originality/value




Despite the importance of OM to big data implementation, for both practitioners and researchers, we have yet to see a systematic analysis of its implementation and its absorptive capacity contribution to building capabilities, at either city system or organizational levels. As such the Markov model makes a preliminary contribution to the literature integrating big data capabilities with OM capabilities and the resulting improvements in system absorptive capacity.

/pdf/exploring-the-influence-of-big-data-on-city-transport-3qvncyqqtt.pdf

Exploring the influence of big data on city transport operations: a Markovian approach

Association rule algorithms can produce a very large number of output patterns. This has raised questions of whether the set of discovered rules "overfit" the data because all the patterns that satisfy some constraints are generated (the Bonferroni effect). In other words, the question is whether some of the rules are "false discoveries" that are not statistically significant. We present a novel approach for estimating the number of "false discoveries" at any cutoff level. Empirical evaluation shows that on typical datasets the fraction of rules that may be false discoveries is very small. A bonus of this work is that the statistical significance measures we compute are a good basis for ordering the rules for presentation to users, since they correspond to the statistical "surprise" of the rule. We also show how to compute confidence intervals for the support and confidence of an association rule, enabling the rule to be used predictively on future data.

/pdf/discovering-predictive-association-rules-5fygm35288.pdf

Byron L. Newton

Papers

Statistics for Business