An Interactive Multi-Task Learning Framework for Next POI Recommendation with Uncertain Check-ins

Studies on next point-of-interest (POI) recommendation mainly seek to learn users' transition patterns with certain historical check-ins. However, in reality, users' movements are typically uncertain (i.e., fuzzy and incomplete) where most existing methods suffer from the transition pattern vanishing issue. To ease this issue, we propose a novel interactive multi-task learning (iMTL) framework to better exploit the interplay between activity and location preference. Specifically, iMTL introduces: (1) temporal-aware activity encoder equipped with fuzzy characterization over uncertain check-ins to unveil the latent activity transition patterns; (2) spatial-aware location preference encoder to capture the latent location transition patterns; and (3) task-specific decoder to make use of the learned latent transition patterns and enhance both activity and location prediction tasks in an interactive manner. Extensive experiments on three real-world datasets show the superiority of iMTL.

Probing enzymatic activity – a radical approach

Latent activity-based probes have been developed for deubiquitinating enzymes using a thiol–ene strategy, labelling following a specific binding interaction.

State-space models to infer movements and behavior of fish detected in a spatial array of acoustic receivers

We developed state-space models for inferring movements and behaviors of fish implanted with acoustic transmitters and detected within a spatial array of stationary acoustic receivers. In these models, fish movements and behavior switching are specified using a hidden Markov model of the changes in an individual’s latent activity center. The observed number of detections at each acoustic receiver is modeled as a function of the distance between the receiver and a fish’s activity center. To illustrate these models, we analyzed the detections of individual Gulf sturgeon (Acipenser oxyrinchus desotoi) that moved within a spatial array of acoustic receivers placed in the Suwannee River, Florida. Our models of these detections provided estimates of the locations of individual sturgeon and the periods when individuals switched between spawning and resting behaviors. We anticipate that these models will be used to analyze acoustic surveys of other species and help design new surveys.

Chilling tolerance in Zizania latifólia

ABSTRACT: The tumescent stems of Zizania latifolia are consumed as vegetable in southern and eastern Asia. This study aimed to compare photosynthesis and chilling tolerance parameters between two well-known Zizania latifolia cultivars: Longjiao 2 and Zhejiao 911, which are chilling tolerant and sensitive, respectively. We found that severe cold stress induced photosynthesis inhibition (5°C) resulted from non-stomatal factors. However, net photosynthetic rate (Pn), stomatal conductance (Gs) and transpiration rate (Tr) of the cultivar Longjiao 2 were higher than that of Zhejiao 911 with more gradual variations. Six parameters of chlorophyll fluorescence including latent activity of PSII (Fv/F0), efficiency of primary photochemistry (Fv/Fm), photochemical quenching coefficient (qP), non-photochemical quenching coefficient (qN), quantum yield of electric transport (Yield) and the ratio of electric transport at PSII (ETR) were analyzed in the two cultivars. We found that Longjiao 2 had significantly higher Fv/Fm and qN than Zhejiao 911 while qP values were only slightly different for the two lines at severe cold stress. In our experiments, Yield and ETR changed in a similar way in both Longjiao 2 and Zhejiao 911. In addition, the ability of heat dissipation of Longjiao 2 was statistically higher than that found in Zhejiao 911 when treated at 5°C for one day. These data suggest that cultivar Longjiao 2 induces chilling tolerance by modulating critical pathways including photosynthesis and energy dissipation.

Biological effects of milk proteins and their peptides with emphasis on those related to the gastrointestinal ecosystem

The protein fraction of milk contains several components with physiological significance for the development of the newborn. Among them, immunomodulatory peptides and lactoferrin exemplify the complexity of biologically active substances of milk. Immunomodulatory peptides have latent activity within the native protein and are generated after proteolysis during gastrointestinal transit. Once they are generated, they modulate mucosal immunity, possibly by guiding the local immune system until it develops its full functionality. Lactoferrin is another milk bioactive compound with nutritional and health promoting properties; it modulates the microbial intestinal environment, displays anti-microbial activity against various pathogens and stimulates the establishment of beneficial microflora. The following overview focuses on the importance of immunomodulatory peptides and lactoferrin for the maturation of intestine and immune system that are functionally immature in the newborn.