Assessing Activity Pattern Similarity with Multidimensional Sequence Alignment Based on a Multiobjective Optimization Evolutionary Algorithm

Abstract

Due to the complexity and multidimensional characteristics of human activities, assessing the similarity of human activity patterns and classifying individuals with similar patterns remains highly challenging. This paper presents a new and unique methodology for evaluating the similarity among individual activity patterns. It conceptualizes multidimensional sequence alignment (MDSA) as a multiobjective optimization problem, and solves this problem with an evolutionary algorithm. The study utilizes sequence alignment to code multiple facets of human activities into multidimensional sequences, and to treat similarity assessment as a multiobjective optimization problem that aims to minimize the alignment cost for all dimensions simultaneously. A multiobjective optimization evolutionary algorithm (MOEA) is used to generate a diverse set of optimal or near-optimal alignment solutions. Evolutionary operators are specifically designed for this problem, and a local search method also is incorporated to improve the search ability of the algorithm. We demonstrate the effectiveness of our method by comparing it with a popular existing method called ClustalG using a set of 50 sequences. The results indicate that our method outperforms the existing method for most of our selected cases. The multiobjective evolutionary algorithm presented in this paper provides an effective approach for assessing activity pattern similarity, and a foundation for identifying distinctive groups of individuals with similar activity patterns.

Space–Time Analysis: Concepts, Quantitative Methods, and Future Directions

“…The usefulness of these methods cannot be overestimated. For instance, comparisons between space-time paths can be conducted through the use of path similarity indexes including Hausdorff distance and Frechet distance, the dynamic time warping algorithm, the multiobjective optimization evolutionary algorithm Probabilistic time geography @ @ Survival analysis models b @ @ Others @ Process-based simulation Agent-based models c @ @ Others (e.g., Markov models) @ Spatial panel data analysis Pattern revelation Multiple space-time metrics @ Multiple space-time tests @ Others @ Space-time statistical models Panel regression models @ S-T autoregressive models & variants @ S-T weighted regression models @ Latent trajectory /multilevel models c @ Survival analysis models c @ @ Others (e.g., hybrid models) @ Process-based simulation Agent-based models @ @ Cellular automaton @ Spatial Markov models @ Others @ a We do not provide detail due to decent coverage in Long and Nelson (2013) Space-Time Analysis: Concepts, Quantitative Methods, and Future Directions (MOEA), and the longest common sequence algorithm (Chen et al 2011;Long and Nelson 2013;Kwan, Xiao, and Ding 2014). Other methods allow for grouping together individuals who have similar spacetime paths or activity density surface (Kwan 1999;Chen et al 2011), quantitatively modeling movement probabilities that incorporate object kinetics (Long, Nelson, and Nathoo 2014), and creating probabilistic space-time prisms depicting an individual agent's daily movement (Downs et al 2014).…”

“…For details about these quantitative methods, see Long and Nelson (2013). Some specific research efforts have focused on development and evaluation of space-time measures or algorithms that are "sensitive to person-specific situations and gender-role constraints" (e.g., Kwan 1998, 211), assessment of the similarity among individual activity patterns using the sequence alignment method that was originally developed to analyze DNA sequences (Shoval and Isaacson 2007;Kwan, Xiao, and Ding 2014), and exploration and visualization of large space-time trajectory data sets in the GIS software environment for both human (e.g., Kwan 2004) and nonhuman agents (e.g., Baer and Butler 2000;Downs et al 2014). In parallel with such efforts, individual movement data analysis also aims at better understanding individual people's activity-travel scheduling behavior subject to space-time prisms (e.g., Liao, Rasouli, and Timmermans 2014) as well as revealing how individual accessibility constraints might factor into personal or social decision making (e.g., Neutens et al 2008).…”

Spatio-Temporal Analytics for Exploring Human Mobility Patterns and Urban Dynamics in the Mobile Age

“…Spatial clustering approaches have been developed for discovering places of interest and similar routes from human movement trajectories (Palma et al, 2008;Giannotti et al, 2011). In addition, sequence alignment method, which was originally employed by biochemists to analyze DNA sequences, has been applied in analyzing the space-time sequential aspects of human activities (Wilson, 1998;Shoval & Isaacson, 2007;Mavoa et al, 2011;Kwan et al, 2014). It could help the identification of individual-based spatio-temporal recurring trends and group-based similarity patterns.…”

Space–Time Sequential Similarity for Identifying Factors of Activity‐Travel Pattern Segmentation: A Measure of Sequence Alignment and Path Similarity

“…The former potentially omits important pattern groups, while the latter does not capture sequential information embedded in the activity implementation along the day. Sequence alignment methods (SAMs) gained ample attention in urban studies and transportation sciences (Wilson 1998;Bargeman, Joh, and Timmermans 2002;Shoval and Isaacson 2007;Huynh, Hall, and Doherty et al 2008;Saneinejad and Roorda 2009;Mavoa, Oliver, and Witten et al 2011;Sammour et al 2012;Kwan, Xiao, and Ding 2015;Chavoshi et al 2015;Dharmowijoyo, Susilo, and Karlström 2017;Xianyu, Rasouli, and Timmermans 2017). The methods can provide the better segmentation of activity-travel patterns that are better associated with the concerned policy variables (Joh, Arentze, Hofman, and Timmermans 2001).…”