The study of the perception of security helps to understand the feeling of citizens in the face of risk and the magnitude of its consequences, to understand the fear of crime. Traditionally the measure of this kind of subjective perception was made through surveys with low costly-effective performance, to a small sample of the population in specific time periods. To solve those inconveniences, this work analyzes take advantage of the amount of data and the real-time monitoring allowed by the social networks to quantifies the Perception of the security in Bogota. The quantification is made through different methods that involve ruled-based and supervised learning approaches for the sentiment analysis of the data coming from Spanish text from Twitter.