University of La Laguna Showcase

This benchmark, based on the Solomon instances, has been generated for solving a Vehicle Routing Problem with Synchronization for Waste Collection. The Solomon instances include information on a single depot and customer locations, demand, time windows, vehicle capacity, and fleet size. To adapt these instances to our multi-synchronization problem, the following modifications were made: up to three waste types are considered simultaneously, with demands assigned using a non-negative discrete uniform distribution bounded above by the original demand values; synchronization time windows are introduced by doubling and halving the original durations; each customer may submit three different bids, with the highest bid assigned to the shortest time window and the lowest to the widest, thereby reflecting a preference for more restrictive time intervals; finally, a fleet of 20 vehicles with homogeneous capacity of 200 units per waste type is assumed. These bid values are computed mathematically based on a baseline corresponding to the cost of serving a customer via the trivial route (depot–customer–depot), to which a small random factor is applied. In cases where transportation costs vary depending on the type of waste, that is, when a function to weigh each type of waste is considered, the same procedure is applied, taking into account the specific waste type generated by each customer. The files storing the instances are denoted as "f", "f2", "ftc", and "f2tc", wherein "tc" indicates an instance created with weighted wastes taken into account and "2" indicates an instance for two types of waste; otherwise, the instances were created for either non-weighted wastes or three types of wastes. Each instance file presents the following variables: — customer: unique ID identifier associated with each customer; 0 is reserved for the depot (numeric). — xCoord: the value of the x coordinate in Euclidean space (numeric). — Coord: the value of the y coordinate in Euclidean space (numeric). — demandWasteType1: non-negative demand value for the first type of waste (numeric). — demandWasteType2: non-negative demand value for the second type of waste (numeric). — demandWasteType3: non-negative demand value for the third type of waste (numeric). — readyTime: value in minutes at which a customer's time window starts; start of the planning horizon if depot (numeric). — dueDateNarrow: value in minutes at which a customer's shortest time window ends (numeric). — dueDate: value in minutes at which a customer's base time window ends (numeric). — dueDateWide: value in minutes at which a customer's widest time window ends (numeric). — service: value in minutes of vehicle's service time at the customer (numeric). — bidNarrow: value associated with the customer's shortest time window (numeric). — bid: value associated with the customer's base time window (numeric). — bidWide: value associated with the customer's widest time window (numeric).

Long term database of morphometric measurements of Blainville’s beaked whales off El Hierro, Canary Islands (Eastern North Atlantic population) from aerial photographs. Includes absolute body length measurements of the whales and proportional measurements of their body at 5% increments from rostrum. For further information please contact arranz@ull.edu.es. University of La Laguna, Tenerife, Spain.

This dataset comprises high-definition video recordings documenting tongue-nibbling behaviour in killer whales (Orcinus orca) observed both under human care at Loro Parque (Orca Ocean, Tenerife, Spain) and in a wild population off the coast of northern Norway. Tongue-nibbling is a rare and understudied tactile interaction with potential affiliative and social bonding functions. The dataset includes: One video file recorded in 2024 in Tverrfjorden (Norway), showing a detailed instance of tongue-nibbling between two wild individuals. Footage captured by Allison Kelly Estevez, and Michael Estevez One video recorded during a ethological study in Loro Parque, illustrating repeated occurrences of the same behaviour in 2013.

Long term database of morphometric measurements of short-finned pilot whales off the Canary Islands (Eastern North Atlantic population) from aerial photographs. Includes absolute body length measurements of the whales and proportional measurements of their body at 5% increments from rostrum. For further information please contact arranz@ull.edu.es. University of La Laguna, Tenerife, Spain.

Raw results from cell and animal experiments

This dataset builds upon the publicly available work by Taranvee, who collected household energy consumption data at one-minute resolution from a smart meter monitoring multiple appliances (e.g., dishwasher, home office, fridge, kitchen). The original release also includes regional weather information (humidity, temperature, atmospheric pressure, etc.), providing a rich contextual layer for understanding how environmental factors influence residential power usage. In this augmented version, we introduce additional consumption columns representing distinct IoT devices, Car charger,Water heater,Air conditioning,Home Theater,Outdoor lights,microwave,Laundry,Pool Pump Each of these new columns tracks an appliance's energy usage in kilowatts ([kW]), effectively broadening the dataset’s scope for modeling complex, multi-device scenarios within a single smart home Original work: https://www.kaggle.com/datasets/taranvee/smart-home-dataset-with-weather-information

This dataset was collected to investigate the hypothesis that individuals with borderline personality disorder (BPD) traits exhibit implicit emotional biases and heightened neural reactivity to negative stimuli compared to healthy controls. The data demonstrates that participants with BPD traits responded faster to negative words in an Implicit Association Test (IAT) and showed distinct neural activity patterns in Event-Related Potentials (ERPs), particularly in the P2, N400, and Late Positive Potential (LPP) components. The dataset includes behavioral measures (reaction times and accuracy) from the IAT, as well as electrophysiological recordings obtained via EEG. Participants were classified into two groups (BPD-vulnerable and healthy controls) based on clinical assessments including the SCID-II, BSL-23, and self-reported history of self-harm. EEG data were collected using a 32-electrode cap and processed to extract ERP amplitudes for congruent and incongruent trials with both positive and negative emotional valence words. The findings can be interpreted to suggest that individuals with BPD traits have an implicit negativity bias and altered cognitive-emotional processing mechanisms. This dataset can be used to explore the relationship between implicit biases, emotional processing, and neural activity, offering potential applications for further research in psychopathology and cognitive neuroscience.

The dataset provides a summary of the experimental results obtained from an HCAI system implemented using a RAG framework and the Llama 3.0 model. A total of 125 hyperparameter configurations were defined by aggregating metrics based on the median of the results from 91 questions and their corresponding answers. These configurations represent the alternatives evaluated through Multi-Criteria Decision-Making (MCDM) methods.

The data reflect the results of the experimentation with an HCAI system implemented using a RAG framework and the Llama 3.0 model. During the experimentation, 91 questions were utilized in the domain of legal advice and migrant rights. Metrics assessed included contextual enrichment, textual quality, discourse analysis, and sentiment evaluation. This allows for the analysis of sentiments and emotions, bias detection, content and toxicity classification, as well as an analysis of inclusion and diversity.

There are new benchmark instances for the Cross-Dock Door Assignment and Scheduling Problem (CDSP), with instances from eight origins and destinations, four inbound and outbound doors, up to 50 origins and destinations, and 30 inbound and outbound doors. The instances have been obtained following the criteria of the paper by Nassief et al. (2016), with some modifications. First of all, the number of pallets to be moved from a supplier to a customer is randomly generated by using a uniform distribution U[1,5] until density values are set to 25, 35, 50, and 75%. Each inbound truck sends pallets from at least one outgoing truck, and each outgoing truck receives pallets from at least one incoming truck. The distance matrix is generated with numbers from the interval [1, 1 + |I|- 1], meaning that a direct distance between two doors is equal to 1, and then an increment of 1 unit is added for the next indirect door. The number of considered incoming/outgoing trucks is 8, 9, 10, 11, 12, 15, 20, and 50. The number of considered inbound/outbound doors is 4, 5, 6, 7, 10, and 30. The door capacities are equal for each instance and calculated by dividing the total flow coming from all origins by the total number of inbound doors and then adding a capacity slackness of 5, 10, 15, 20, and 30%. To generate this set of instances, we have considered the combinations of values reported in the paper by Sayed (2020). The instances are referred to as 00×00×00x00 (number of the incoming and outgoing trucks x number of inbound and outbound doors x capacity slackness associated with the inbound/outbound doors - 5, 10, 15, 20, and 30% - x density of the flow matrix - 25, 35, 50, and 75%).