Adaptation of a Clustering Algorithm and Mosquito Swarm to a problem of ovitraps for the Dengue Mosquito Vector María Beatríz Bernábe-Loranca 1 , Marco Antonio Rodríguez-Flores 2 , Ruth Aralí Martínez-Vega 3 , José Ramos- Castañeda 4 , Elias Olivares- Benitez 5 1,2 Benemérita Universidad Autónoma de Puebla BUAP, Facultad de Ciencias de la Computación 1,5 Universidad Popular Autónoma del Estado de Puebla UPAEP 3,4 Instituto Nacional de Salud Pública (INSP) Abstract—The efforts to locate the ovitraps in an homogenous way in a determined community must be redouble still, the reality is that the location of these depends too much of the will of the community and of the use that is given to the piece of land where the ovitrap should be located. Given that around the spot of the dengue cases the transmission is more probable (infected mosquitos) and that it’s been documented that vertical transmission exists (adult mosquito to eggs), to the viral and entomological vigilance of endemic communities is important to know and to monitor, through the research of the material obtained by the ovitraps, the strains of Dengue virus that circulate between the human population and the mosquito populations. If the ovitraps were located according to a random representative design and homogenous in the community of study, the vigilance system described above would consist in studying the ovitraps in the range of flight of the mosquito (200 meters approximately). However, in reality this is not the case, therefore a probabilistic approximation is required to establish which ovitraps should be evaluated by the system of sanitary vigilance to have a higher probability of success in the diagnosis, making the process cost-efficient. In this scenery, in accordance to the mobility of the mosquito, a clustering algorithm has been associated and adapted based in P-means that promises to construct groups where the center of each group is a case and the closest ovitraps are associated to establish a systematic and homogenous configuration of the relationship between a registered case and the ovitraps. Keywords-Cluster, Cases, Dengue, Mosquito, Ovitraps I. INTRODUCTION The system of entomological vigilance that is carried out by the authorities of national health is based in the called ovitraps; these are the units were the Aedes mosquito females lay eggs after feeding with human blood. In this point, it is assumed that the number of eggs in said devices is proportional to the number of vectors that transmit Dengue virus. All Vector control programs, which have focused mainly on the patient house and peridomestic areas around dengue cases, have not produced the expected impact on transmission. To evaluate the assumption that the endemic/epidemic transmission of dengue begins around peridomestic vicinities of the index dengue cases, a prospective cohort study was conducted (in Tepalcingo and Axochiapan, in the state of Morelos, Mexico), using the state surveillance system for the detection of incident cases. Paired blood specimens were collected from both the individuals who live with the incident cases and a sample of subjects residing within a 25- meter radius of such cases (exposed cohort), in order to measure dengue-specific antibodies. Other subjects were selected from areas which have not presented any incident cases within 200 meters, during the two months preceding the sampling (non-exposed cohort) [1]. Symptomatic/asymptomatic incident infection was detected. In the analysis it will be considered as the dependent variable, the exposure to confirmed dengue cases as the main independent variable, and the estimated Ae. Aegypti abundance and their infection with DENV, also socio-demographic and socio-cultural conditions of the subjects will be considered as additional explanatory variables [1]. As it was impossible to monitor and ascertain vector density in the domiciles of the participating subjects prior to and during follow up, we want to estimate this variable based in the data from the ovitrap-based surveillance program. Also, due to the difficult logistics involved in the capture of adult Aedes, the mosquito infection will be analyzed in the individuals that emerge from the eggs collected through the ovitraps closest to the subjects’ dwellings, but we don’t know how to select these ovitraps because their location into the endemic area is not homogeneous. To answer to this situation, we propose configuring a structure of the ovitraps under schemes of clustering algorithms analogue to the swarms of mosquitos and in particular to the sample of the behavior of the dengue mosquito. The study in this work starts with the introduction as section 1. In the section 2 we expose the way of life of the swarms of mosquitos as a bioinspired algorithm. We continue the section 3 with a brief state of the art of the problem of the Dengue mosquito giving place to describe in section 4, a proposal of algorithmic solution of clustering with analogy to the life of swarms of mosquitos and finally in section 5 we present a discussion of the results and lastly the conclusions. II. MOSQUITO SWARM ALGORITHM We can see on the literature a new propose classification of meta-heuristics algorithms not based on swarm intelligence theory but rather on grouping of animals: swarm algorithms, schools algorithms, flocks algorithms and herds 108 Trends in Innovative Computing 2012 - Nature Inspired Computing