Hybrid Weighted Barebones Exploiting Particle Swarm Optimization Algorithm for Time Series Representation Antonio Manuel Dur´ an-Rosal (B) , David Guijo-Rubio, Pedro Antonio Guti´errez, and C´esarHerv´as-Mart´ınez Department of Computer Science and Numerical Analysis, University of C´ordoba, C´ordoba, Spain i92duroa@uco.es Abstract. The amount of data available in time series is recently increasing in an exponential way, making difficult time series prepro- cessing and analysis. This paper adapts different methods for time series representation, which are based on time series segmentation. Specifically, we consider a particle swarm optimization algorithm (PSO) and its bare- bones exploitation version (BBePSO). Moreover, a new variant of the BBePSO algorithm is proposed, which takes into account the positions of the particles throughout the generations, where those close in time are given more importance. This methodology is referred to as weighted BBePSO (WBBePSO). The solutions obtained by all the algorithms are finally hybridised with a local search algorithm, combining simple seg- mentation strategies (Top-Down and Bottom-Up). WBBePSO is tested in 13 time series and compared against the rest of algorithms, showing that it leads to the best results and obtains consistent representations. Keywords: Time series representation · Segmentation Barebones particle swarm optimization · Hybrid algorithms 1 Introduction Nowadays, the exponential increase of time series and their big amount of data hamper their processing [1]. Time series data mining (TSDM) includes several tasks such as the reconstruction of missing values [2], clustering [3], classification [4], forecasting [5] or segmentation [6]. Different areas of application can signifi- cantly benefit from efficient TSDM algorithms, including climate [7] or finances [8], among others. Time series segmentation consists in dividing the time series into consecutive parts or points, trying to satisfy different objectives. There are two points of view that time series segmentation is focused on. On the one hand, segmenting time series is used for discovering patterns in them. On the other hand, there is c  Springer International Publishing AG, part of Springer Nature 2018 P. Koroˇsec et al. (Eds.): BIOMA 2018, LNCS 10835, pp. 126–137, 2018. https://doi.org/10.1007/978-3-319-91641-5_11