Discretization of Time Series Data

Abstract

An increasing number of algorithms for biochemical network inference from experimental data require discrete data as input. For example, dynamic Bayesian network methods and methods that use the framework of finite dynamical systems, such as Boolean networks, all take discrete input. Experimental data, however, are typically continuous and represented by computer floating point numbers. The translation from continuous to discrete data is crucial in preserving the variable dependencies and thus has a significant impact on the performance of the network inference algorithms. We compare the performance of two such algorithms that use discrete data using several different discretization algorithms. One of the inference methods uses a dynamic Bayesian network framework, the other-a time-and state-discrete dynamical system framework. The discretization algorithms are quantile, interval discretization, and a new algorithm introduced in this article, SSD. SSD is especially designed for short time series data and is capable of determining the optimal number of discretization states. The experiments show that both inference methods perform better with SSD than with the other methods. In addition, SSD is demonstrated to preserve the dynamic features of the time series, as well as to be robust to noise in the experimental data. A C++ implementation of SSD is available from the authors at http://polymath.vbi.vt.edu/discretization .

A Feature Extraction Method for Multivariate Time Series Classification Using Temporal Patterns

“…For reducing and simplifying the original data, numerous values of a continuous variable is always be replaced by a small number of interval labels, which leads to a concise, easy-to-use, knowledge-level representation of mining results [22]. Data discretization is a frequently used technique to partition the value space of a continuous variable into a finite number of intervals and assigning a nominal value of each of them [23,24]. Equal Width and Equal Frequency are two simplest discretization methods.…”

Boolean monomial control systems

“…Since PathSim is a stochastic agentiii based computer simulation, Dr. Laubenbacher's idea was to use the average output of PathSim as data to construct (or to "reverse engineer") a deterministic, time discrete dynamical system over a suitable finite field. To this end, he and some of his graduate students had developed some specific methods 1 [65], [35]. This approach needed to be analyzed and tested, so, I performed the mathematical analysis of the reverse engineering method described in [65].…”

A Harmony Search Algorithm with Multi-pitch Adjustment Rate for Symbolic Time Series Data Representation

“…Many studies have proposed improvements to SAX. These studies [9,[14][15][16][17][18] have shown that the SAX method is still an open field of research and that new developments are required to bring new ideas to improve the performance of SAX.…”

“…This method processes a single time series at a time, so the discretisation criterion is not generalized to the complete dataset. In [16], time series values are represented as a multi-connected graph. Using this representation, similar time series are grouped in a graphical model.…”