This research explores a novel Mexican Sign Language (MSL) lexicon video dataset containing the dynamic gestures most frequently used in MSL. Each gesture consists of a set of different versions of videos under uncontrolled conditions. The MX-ITESO-100 dataset is composed of a lexicon of 100 gestures and 5000 videos from three participants with different grammatical elements. Additionally, the dataset is evaluated in a two-step neural network model as having an accuracy greater than 99% and thus serves as a benchmark for future training of machine learning models in computer vision systems. Finally, this research provides an inclusive environment within society and organizations, in particular for people with hearing impairments.

In the real world, the assumption that, once generated, a pervasive system of systems and the services it provides will remain static is false. Systems can leave or enter, and service availability in systems can change along with the environmental conditions. The dynamic composition and adaptation of pervasive systems of systems enable them to expand their functionality through leverage resources in the user vicinity. Most of the adaptation approaches based on ad hoc networks use service adaptation created from scratch. That is, each time the environment changes, the system starts the composition process over again, resulting in the depletion of system resources and network capacity. We present a protocol for dynamic composition of a pervasive system of systems and their services in ad hoc networks. Our proposal uses a dynamic and distributed constraint satisfaction approach to establish the pervasive-system-of-systems requirements. In this paper, we deal with spatial and temporal constraints. However, any other kind of restriction, for instance, quality, can be addressed in the same manner. Additionally, a heuristic allowing recomposing services based on the identification of service components disqualified due to the dynamism of the environment is presented. Thus, the adaptation of services is not necessarily done from scratch. Finally, we present simulations that show the performance of our proposal regarding the nearest work in the state of the art, both in time and in number of messages consumed for adapting services.

Graph mining has emerged as a significant field of research with applications spanning multiple domains, including marketing, corruption analysis, business, and politics. The exploration of knowledge within graphs has garnered considerable attention due to the exponential growth of graph-modeled data and its potential in applications where data relationships are a crucial component, and potentially being even more important than the data themselves. However, the increasing use of graphs for data storing and modeling presents unique challenges that have prompted advancements in graph mining algorithms, data modeling and storage, query languages for graph databases, and data visualization techniques. Despite there being various methodologies for data analysis, they predominantly focus on structured data and may not be optimally suited for highly connected data. Accordingly, this work introduces a novel methodology specifically tailored for knowledge discovery in labeled and heterogeneous graphs (KDG), and it presents three case studies demonstrating its successful application in addressing various challenges across different application domains.

Knowledge Graphs (KGs) have emerged as a powerful tool for representing semantic structured information and enabling the development of intelligent systems. This paper focuses on the generation of semantic maps as summarization method for KGs. We propose a strategy that utilizes centroid-based clustering algorithms, namely Affinity Propagation and Partitioning Around Medoids (PAM), to capture the semantic distance between nodes in the KG and generate meaningful clusters. Our experiments demonstrate divergent results between the two clustering algorithms, with Affinity Propagation showing qualitative coherence and meaningfulness, while PAM performs well in terms of internal validation metrics. We leverage the computed centroids to infer a main term of the semantic map, which contributes to the visually appealing and informative representation of the KG. The combination of semantic distance capture, clustering algorithms, and centroid-based inference facilitates a comprehensive understanding of the KG. Our findings highlight the importance of considering both qualitative and quantitative evaluation measures in assessing clustering results. The effectiveness of semantic maps is showcased in visualizing KGs and advancing the field of knowledge graph visualization. The integration of centroid-based clustering algorithms, qualitative evaluation, and inference methods offers improved clarity and interpretability for complex KG analysis.

In Mexico, the incorporation of deaf people into education has been lacking since only 14% of the deaf population in the age group between 3 and 29 years access education with the support of a hearing aid. Additionally, those who have been incorporated frequently face inappropriate educational strategies which poorly develop the use of Mexican Sign Language (MSL) and therefore academical success and opportunities for insertion in the workplace are difficult. This research explores a novel mexican sign language lexicon video dataset containing the dynamical gestures most frequently used by MSL. Each gesture consists of a set of different versions of videos under uncontrolled conditions. MX-ITESO-100 data set is composed of a lexicon of 100 gestures and 5,000 videos from three participants with different grammatical elements. Additionally, the data set is evaluated in a two-step neural network model with an accuracy greater than 99%. and thus serves as a benchmark for future training of machine learning models in computer vision systems. Finally, this research provides an inclusive environment within society and organizations in particular for people with hearing impairment.

Anomaly-based intrusion detection systems use profiles to characterize expected behavior of network users. Most of these systems characterize the entire network traffic within a single profile. This work proposes a user-level anomaly-based intrusion detection methodology using only the user’s network traffic. The proposed profile is a collection of TopK rankings of reached services by the user. To detect unexpected behaviors, the real-time traffic is organized into TopK rankings and compared to the profile using similarity measures. The experiments demonstrated that the proposed methodology was capable of detecting a particular kind of malware attack in all the users tested.