RESUMO
A relevant goal in human-computer interaction is to produce applications that are easy to use and well-adjusted to their users' needs. To address this problem it is important to know how users interact with the system. This work constitutes a methodological contribution capable of identifying the context of use in which users perform interactions with a groupware application (synchronous or asynchronous) and provides, using machine learning techniques, generative models of how users behave. Additionally, these models are transformed into a text that describes in natural language the main characteristics of the interaction of the users with the system.
RESUMO
Imagine an agent that performs tasks according to different strategies. The goal of Behavioral Recognition (BR) is to identify which of the available strategies is the one being used by the agent, by simply observing the agent's actions and the environmental conditions during a certain period of time. The goal of Behavioral Cloning (BC) is more ambitious. In this last case, the learner must be able to build a model of the behavior of the agent. In both settings, the only assumption is that the learner has access to a training set that contains instances of observed behavioral traces for each available strategy. This paper studies a machine learning approach based on Probabilistic Finite Automata (PFAs), capable of achieving both the recognition and cloning tasks. We evaluate the performance of PFAs in the context of a simulated learning environment (in this case, a virtual Roomba vacuum cleaner robot), and compare it with a collection of other machine learning approaches.
RESUMO
Colorectal cancer (CRC) is the second most deadly and the third most diagnosed cancer in both sexes worldwide. CRC pathogenesis is associated with risk factors such as genetics, alcohol, smoking, sedentariness, obesity, unbalanced diets, and gut microbiota dysbiosis. The gut microbiota is the microbial community living in symbiosis in the intestine, in a dynamic balance vital for health. Increasing evidence underscores the influence of specific gut microbiota bacterial species on CRC incidence and pathogenesis. In this regard, conjugated linoleic acid (CLA) metabolites produced by certain gut microbiota have demonstrated an anticarcinogenic effect in CRC, influencing pathways for inflammation, proliferation, and apoptosis. CLA production occurs naturally in the rumen, and human bioavailability is through the consumption of food derived from ruminants. In recent years, biotechnological attempts to increase CLA bioavailability in humans have been unfruitful. Therefore, the conversion of essential dietary linoleic acid to CLA metabolite by specific intestinal bacteria has become a promising process. This article reviews the evidence regarding CLA and CLA-producing bacteria as therapeutic agents against CRC and investigates the best strategy for increasing the yield and bioavailability of CLA. Given the potential and limitations of the present strategies, a new microbiome-based precision nutrition approach based on endogenous CLA production by human gut bacteria is proposed. A literature search in the PubMed and PubMed Central databases identified 794 papers on human gut bacteria associated with CLA production. Of these, 51 studies exploring association consistency were selected. After excluding 19 papers, due to health concerns or discrepancies between studies, 32 papers were selected for analysis, encompassing data for 38 CLA-producing bacteria, such as Bifidobacterium and Lactobacillus species. The information was analyzed by a bioinformatics food recommendation system patented by our research group, Phymofood (EP22382095). This paper presents a new microbiome-based precision nutrition approach targeting CLA-producing gut bacterial species to maximize the anticarcinogenic effect of CLA in CRC.
RESUMO
INTRODUCTION: There is a need for biomarkers to monitor the earliest phases of Parkinson's disease (PD), especially in premotor stages. Here, we studied whether there are early gait alterations in carriers of the G2019S mutation of LRRK2 that can be detected by means of an inertial sensor system. METHODS: Twenty-one idiopathic PD patients, 20 LRRK2-G2019S PD, 27 asymptomatic carriers of LRRK2-G2019S mutation (AsG2019S) and 36 controls walked equipped with 16 lightweight inertial sensors in three different experiments: i/normal gait, ii/fast gait and iii/dual-task gait. In the AsG2019S group, DaT-SPECT (123I-ioflupane) with semi-quantitative analysis was carried out. Motor and cognitive performance were evaluated using MDS-UPDRS-III and MoCA scales. We employed neural network techniques to classify individuals based on their walking patterns. RESULTS: PD patients and controls showed differences in speed, stride length and arm swing amplitude, variability and asymmetry in all three tasks (p < 0.01). In the AsG2019S group, the only differences were detected during fast walking, with greater step time on the non-dominant side (p < 0.05), lower step/stride time variability (p < 0.01) and lower step time asymmetry (p < 0.01). DaT uptake showed a significant correlation with step time during fast walking on the non-dominant side (r = -0.52; p < 0.01). The neural network was able to differentiate between AsG2019S and healthy controls with an accuracy rate of 82.5%. CONCLUSION: Our sensor-based analysis did not detect substantial and robust changes in the gait of LRRK2-G2019S asymptomatic mutation carriers. Nonetheless, step or stride time during fast walking, supported by the observed correlation with striatal DaT binding deserves consideration as a potential biomarker in future studies.