Session Recommendation Model Based on Context-Aware and Gated Graph Neural Networks

The graph neural network (GNN) based approach has been successfully applied to session-based recommendation tasks. However, in the face of complex and changing real-world situations, the existing session recommendation algorithms do not fully consider the context information in user decision-making; furthermore, the importance of context information for the behavior model has been widely recognized. Based on this, this paper presents a session recommendation model based on context-aware and gated graph neural networks (CA-GGNNs). First, this paper presents the session sequence as data of graph structure. Second, the embedding vector representation of each item in the session graph is obtained by using the gated graph neural network (GGNN). In this paper, the GRU in GGNN is expanded to replace the input matrix and the state matrix in the conventional GRU with input context captured in the session (e.g., time, location, and holiday) and interval context (representing the proportion of the total session time of each item in the session). Finally, a soft attention mechanism is used to capture users’ interests and preferences, and a recommendation list is given. The CA-GGNN model combines session sequence information with context information at each time. The results on the open Yoochoose and Diginetica datasets show that the model has significantly improved compared with the latest session recommendation methods.

Performing Repetitions To Failure in Lower-Limb Single-Joint Exercise does not Reduce Countermovement Jump Performance in Trained Male Adults

Performing repetitions to failure (RF) is a strategy that might acutely reduce neuromuscular performance, as well as increase the rating of perceived exertion (RPE) and the internal training load (ITL) during and after a resistance training (RT) session. Thus, this study aimed to analyze the acute effects of RF or repetitions not to failure (RNF) on countermovement jump (CMJ) performance and the ITL in trained male adults. Eleven men performed two experimental protocols in randomized order (RF vs. RNF). Under the RF condition, participants performed three sets of the leg extension exercise using 100% of the 10RM load and rest intervals of 180-s between sets. Under the RNF condition, participants were submitted to six sets of five repetitions with the same intensity and an 80-s rest interval between sets in the same exercise. The CMJ test was analyzed before and following (15-s and 30-min, respectively) each experimental session. The ITL was evaluated by multiplying the RPE and the total session time, 30-min after the protocol. No main effect or interaction time vs. condition was found for CMJ performance (p > 0.05). In contrast, the ITL showed higher values under the RF condition (p = 0.003). Therefore, even though RF-induced a greater ITL, our results suggest that adopting this strategy in one single-joint exercise for the lower limbs does not seem sufficient to reduce CMJ height.

Group activity with Paro in nursing homes: systematic investigation of behaviors in participants

ABSTRACTBackground:A variety of group activities is promoted for nursing home (NH) residents with dementia with the aim to reduce apathy and to increase engagement and social interaction. Investigating behaviors related to these outcomes could produce insights into how the activities work. The aim of this study was to systematically investigate behaviors seen in people with dementia during group activity with the seal robot Paro, differences in behaviors related to severity of dementia, and to explore changes in behaviors.Methods:Thirty participants from five NHs formed groups of five to six participants at each NH. Group sessions with Paro lasted for 30 minutes twice a week during 12 weeks of intervention. Video recordings were conducted in the second and tenth week. An ethogram, containing 18 accurately defined and described behaviors, mapped the participants’ behaviors. Duration of behaviors, such as “Observing Paro,” “Conversation with Paro on the lap,” “Smile/laughter toward other participants,” were converted to percentage of total session time and analyzed statistically.Results:“Observing Paro” was observed more often in participants with mild to moderate dementia (p = 0.019), while the variable “Observing other things” occurred more in the group of severe dementia (p = 0.042). “Smile/laughter toward other participants” showed an increase (p = 0.011), and “Conversations with Paro on the lap” showed a decrease (p = 0.014) during the intervention period.Conclusions:Participants with severe dementia seemed to have difficulty in maintaining attention toward Paro during the group session. In the group as a whole, Paro seemed to be a mediator for increased social interactions and created engagement.