Marco Antonio Moreno Ibarra

Assessment of the artificial habitat in shrimp aquaculture using environmental pattern classification

This paper presents a novel model for assessing the water quality for the artificial habitat in shrimp aquaculture. The physical-chemical variables involved in the artificial habitat are measured and studied for modeling the environment of the ecosystem. A new physical-chemical index (Γ) classifies the behavior of the environmental variables, calculating the frequency and the deviations of the measurements based on impact levels. A fuzzy inference system (FIS) is used for establishing a relationship between environmental variables, describing the negative ecological impact of the concentrations reported. The FIS uses a reasoning process for classifying the environmental levels providing a new index, which describes the general status of the water quality (WQI); excellent, good, regular and poor.

Semantic Recommender System for Touristic Context Based on Linked Data

The lack of personalization presented in touristic itineraries that are offered by travel agencies involve a little flexibility. Basically, they are designed with the points of interest (POIs) that have more relevance in the area. On the other hand, there are POIs that have agreements with the agencies, which originate a excluding POIs that could be interesting for the tourist. In this work, a method capable to use the user preferences, like POIs and activities that user wants to realize during their vacations is proposed. Moreover, some weighted features such as the max distance that user wants to walk between POIs, and opinions of other users, coming from the web 2.0 by means of social media are taken into account. As result, a personalized route, which is composed of recommended POIs for the user and satisfied the user profile is provided.

Using Soft Similarity in Multi-label Classification for Reuters-21578 Corpus

In classification tasks one of the main problems is to choose which features provide best results, i.e., Construct a vector space model. In this paper, we show how to complement traditional vector space model with the concept of soft similarity. We use the combination of the traditional tf-idf model with latent Dirichlet allocation applied in multi-label classification. We considered multi-label files of the Reuters-21578 corpus as study case. The methodology is evaluated using the multi-label algorithm Rakell. We used the traditional tf-idf model as the baseline. We present the F1 measures for both models for various feature sets, preprocessing techniques and vector sizes. The new model obtains better results than the base line model.

Geocoding of microblogs based on ontologies

This paper proposes a methodology for geocoding microblogs texts in Spanish, in order to identify the location of entities, spatial relationships and events pertaining to a particular phenomenon, for disambiguation and verification of consistency based on ontologies in order to provide certainty in the identified locations.