Browsing by Keyword "Clustering"
Now showing 1 - 6 of 6
Results Per Page
Sort Options
Item Graph Based Learning for Building Prediction in Smart Cities(2022-04) Garmendia-Orbegozo, Asier; Noye, Sarah; Anton, Miguel Angel; Nunez-Gonzalez, J. David; Tecnalia Research & Innovation; DIGITALIZACIÓN Y AUTOMATIZACIÓN DE LA CONSTRUCCIÓNAnticipating pedestrians’ activity is a necessary task for providing a safe and energy efficient environment in an urban area. By locating strategically sensors throughout the city useful information could be obtained. By knowing the average activity of those throughout different days of the week we could identify the typology of the buildings neighboring those sensors. For these type of purposes, clustering methods show great capability forming groups of items that have great similarity intra clusters and dissimilarity inter cluster. Different approaches are made to classify sensors depending on the typology of buildings surrounding them and the mean pedestrians’ counts for different time intervals. By this way, sensors could be classified in different groups according to their activation patterns and the environment in which they are located through clustering processes and using graph convolutional networks. This study reveals that there is a close relationship between the activity pattern of the pedestrians’ and the type of environment sensors that collect pedestrians’ data are located. By this way, institutions could alleviate a great amount of effort needed to ensure safe and energy efficient urban areas, only knowing the typology of buildings of an urban zone.Item An intelligent decision support system for assessing the default risk in small and medium-sized enterprises(Springer Verlag, 2017) Manjarres, Diana; Landa-Torres, Itziar; Andonegui, Imanol; Zurada, Jacek M.; Zadeh, Lotfi A.; Tadeusiewicz, Ryszard; Rutkowski, Leszek; Korytkowski, Marcin; Scherer, Rafal; IA; Tecnalia Research & InnovationIn the last years, default prediction systems have become an important tool for a wide variety of financial institutions, such as banking systems or credit business, for which being able of detecting credit and default risks, translates to a better financial status. Nevertheless, small and medium-sized enterprises did not focus its attention on customer default prediction but in maximizing the sales rate. Consequently, many companies could not cope with the customers’ debt and ended up closing the business. In order to overcome this issue, this paper presents a novel decision support system for default prediction specially tailored for small and medium-sized enterprises that retrieves the information related to the customers in an Enterprise Resource Planning (ERP) system and obtain the default risk probability of a new order or client. The resulting approach has been tested in a Graphic Arts printing company of The Basque Country allowing taking prioritized and preventive actions with regard to the default risk probability and the customer’s characteristics. Simulation results verify that the proposed scheme achieves a better performance than a naïve Random Forest (RF) classification technique in real scenarios with unbalanced datasets.Item A novel grouping harmony search algorithm for clustering problems(Springer Verlag, 2017) Landa-Torres, Itziar; Manjarres, Diana; Gil-López, Sergio; Del Ser, Javier; Sanz, Sancho Salcedo; Del Ser, Javier; Tecnalia Research & Innovation; IAThe problem of partitioning a data set into disjoint groups or clusters of related items plays a key role in data analytics, in particular when the information retrieval becomes crucial for further data analysis. In this context, clustering approaches aim at obtaining a good partition of the data based on multiple criteria. One of the most challenging aspects of clustering techniques is the inference of the optimal number of clusters. In this regard, a number of clustering methods from the literature assume that the number of clusters is known a priori and subsequently assign instances to clusters based on distance, density or any other criterion. This paper proposes to override any prior assumption on the number of clusters or groups in the data at hand by hybridizing the grouping encoding strategy and the Harmony Search (HS) algorithm. The resulting hybrid approach optimally infers the number of clusters by means of the tailored design of the HS operators, which estimates this important structural clustering parameter as an implicit byproduct of the instance-to-cluster mapping performed by the algorithm. Apart from inferring the optimal number of clusters, simulation results verify that the proposed scheme achieves a better performance than other naïve clustering techniques in synthetic scenarios and widely known data repositories.Item A Novel Heuristic Approach for the Simultaneous Selection of the Optimal Clustering Method and Its Internal Parameters for Time Series Data(Springer Verlag, 2020) Navajas-Guerrero, Adriana; Manjarres, Diana; Portillo, Eva; Landa-Torres, Itziar; Martínez Álvarez, Francisco; Troncoso Lora, Alicia; Sáez Muñoz, José António; Corchado, Emilio; Quintián, Héctor; IAClustering methods have become popular in the last years due to the need of analyzing the high amount of collected data from different fields of knowledge. Nevertheless, the main drawback of clustering is the selection of the optimal method along with its internal parameters in an unsupervised environment. In the present paper, a novel heuristic approach based on the Harmony Search algorithm aided with a local search procedure is presented for simultaneously optimizing the best clustering algorithm (K-means, DBSCAN and Hierarchical clustering) and its optimal internal parameters based on the Silhouette index. Extensive simulation results show that the presented approach outperforms the standard clustering configurations and also other works in the literature in different Time Series and synthetic databases.Item Novel hybrid heuristics for an extension of the dynamic relay deployment problem over disaster areas(2014-10) Bilbao, Miren Nekane; Gil-López, Sergio; Del Ser, Javier; Salcedo-Sanz, Sancho; Sánchez-Ponte, Mikel; Arana-Castro, Antonio; IA; GENERALIn this paper, we propose a novel autonomous intelligent tool for the optimum design of a wireless relayed communication network deployed over disaster areas. The so-called dynamic relay deployment problem consists of finding the optimum number of deployed relays and their location aimed at simultaneously maximizing the overall number of mobile nodes covered and minimizing the cost of the deployment. In this paper, we extend the problem by considering diverse relay models characterized by different coverage radii and associated costs. To efficiently tackle this problem we derive a novel hybrid scheme comprising: (1) a Harmony Search (HS)-based global search procedure and (2) a modified version of the well-known K-means clustering algorithm as a local search technique. Single- and bi-objective formulations of the algorithm are proposed for emergency and strategic operational planning, respectively. Monte Carlo simulations are run over a emulated scenario based on real statistical data from the Castilla La Mancha region (center of Spain) to show that, in comparison with a standard implementation of the K-means algorithm followed by a exhaustive search procedure over all relay-model combinations, the proposed scheme renders on average better coverage levels and reduced costs providing, at the same time, an intelligent tool capable of simultaneously determining the number and models of the relays to be deployed.Item Unsupervised methods for anomalies detection through intelligent monitoring systems(2009) Carrascal, Alberto; Díez, Alberto; Azpeitia, Ander; Tecnalia Research & InnovationThe success of intelligent diagnosis systems normally depends on the knowledge about the failures present on monitored systems. This knowledge can be modelled in several ways, such as by means of rules or probabilistic models. These models are validated by checking the system output fit to the input in a supervised way. However, when there is no such knowledge or when it is hard to obtain a model of it, it is alternatively possible to use an unsupervised method to detect anomalies and failures. Different unsupervised methods (HCL, K-Means, SOM) have been used in present work to identify abnormal behaviours on the system being monitored. This approach has been tested into a real-world monitored system related to the railway domain, and the results show how it is possible to successfully identify new abnormal system behaviours beyond those previously modelled well-known problems.