Documenting tacit knowledge in organizations can be a challenging task due to incomplete initial information, difficulty in identifying knowledgeable individuals, the interplay of formal hierarchies and informal networks, and the need to ask the right questions. To address this, we propose an agent-based framework leveraging large language models (LLMs) to iteratively reconstruct dataset descriptions through interactions with employees. Modeling knowledge dissemination as a Susceptible-Infectious (SI) process with waning infectivity, we conduct 864 simulations across various synthetic company structures and different dissemination parameters. Our results show that the agent achieves 94.9% full-knowledge recall, with self-critical feedback scores strongly correlating with external literature critic scores. We analyze how each simulation parameter affects the knowledge retrieval process for the agent. In particular, we find that our approach is able to recover information without needing to access directly the only domain specialist. These findings highlight the agent’s ability to navigate organizational complexity and capture fragmented knowledge that would otherwise remain inaccessible.

With its vast territory and diverse population distribution, Brazil is undergoing significant developments in its energy sector as it transitions towards a liberalized market. By 2030, the country aims to open the market to all consumers, enhancing competition, and liquidity in energy contracts, and fostering a dynamic trading environment. Effective market price monitoring is essential for understanding its dynamics and guiding decision-making processes in engineering applications. This paper aims to predict medium and long-term price movements in the free energy market. Our analysis of market price curves for different time horizons offers insights into energy market dynamics, enabling energy companies to optimize portfolio strategies and mitigate risks effectively. We consider forecast horizons of one quarter and one semester ahead, providing valuable insights for stakeholders navigating Brazil's evolving energy landscape. We utilize a combination of Machine Learning (ML)-based forecasters enriched by a broad set of exogenous data sources to provide forecasts for energy contracts whose effective date is Y2, Y3, and Y4 years ahead. In this study, we explore a simple yet effective and explainable alternative to using all the exogenous time series directly as input for our models. Namely, multiple related time series are aggregated into a series of histograms. Our proposed ensemble outperforms traditional statistical-based forecasting models, other ML solutions, and strong baselines in terms of resulting financial returns, particularly in the one-quarter prediction horizon, while also achieving competitive results for one semester ahead.

Predictive Maintenance applications are increasingly complex, with interactions between many components. Black-box models are popular approaches based on deep-learning techniques due to their predictive accuracy. This paper proposes a neural-symbolic architecture that uses an online rule-learning algorithm to explain when the black-box model predicts failures. The proposed system solves two problems in parallel: (i) anomaly detection and (ii) explanation of the anomaly. For the first problem, we use an unsupervised state-of-the-art autoencoder. For the second problem, we train a rule learning system that learns a mapping from the input features to the autoencoder’s reconstruction error. Both systems run online and in parallel. The autoencoder signals an alarm for the examples with a reconstruction error that exceeds a threshold. The causes of the signal alarm are hard for humans to understand because they result from a non-linear combination of sensor data. The rule that triggers that example describes the relationship between the input features and the autoencoder’s reconstruction error. The rule explains the failure signal by indicating which sensors contribute to the alarm and allowing the identification of the component involved in the failure. The system can present global explanations for the black box model and local explanations for why the black box model predicts a failure. We evaluate the proposed system in a real-world case study of Metro do Porto and provide explanations that illustrate its benefits.

Data production has followed an increased growth in the last years, to the point that traditional or batch machine-learning (ML) algorithms cannot cope with the sheer volume of generated data. Stream or online ML presents itself as a viable solution to deal with the dynamic nature of streaming data. Besides coping with the inherent challenges of streaming data, online ML solutions must be accurate, fast, and bear a reduced memory footprint. We propose a new decision tree-based ensemble algorithm for online ML regression named online extra trees (OXT). Our proposal takes inspiration from the batch learning extra trees (XT) algorithm, a popular and faster alternative to random forest (RF). While speed and memory costs might not be a central concern in most batch applications, they become crucial in data stream data learning. Our proposal combines subbagging (sampling without replacement), random tree split points, and model trees to deliver competitive prediction errors and reduced computational costs. Throughout an extensive experimental evaluation comprising 22 real-world and synthetic datasets, we compare OXT against the state-of-the-art adaptive RF (ARF) and other incremental regressors. OXT is generally more accurate than its competitors while running significantly faster than ARF and expending significantly less memory.

River is a machine learning library for dynamic data streams and continual learning. It provides multiple state-of-the-art learning methods, data generators/transformers, performance metrics and evaluators for different stream learning problems. It is the result from the merger of two popular packages for stream learning in Python: Creme and scikit-multiflow. River introduces a revamped architecture based on the lessons learnt from the seminal packages. River's ambition is to be the go-to library for doing machine learning on streaming data. Additionally, this open source package brings under the same umbrella a large community of practitioners and researchers. The source code is available at https://github.com/online-ml/river.

A central aspect of online decision tree solutions is evaluating the incoming data and enabling model growth. For such, trees much deal with different kinds of input features and partition them to learn from the data. Numerical features are no exception, and they pose additional challenges compared to other kinds of features, as there is no trivial strategy to choose the best point to make a split decision. The problem is even more challenging in regression tasks because both the features and the target are continuous. Typical online solutions evaluate and store all the points monitored between split attempts, which goes against the constraints posed in real-time applications. In this paper, we introduce the Quantization Observer (QO), a simple yet effective hashing-based algorithm to monitor and evaluate split point candidates in numerical features for online tree regressors. QO can be easily integrated into incremental decision trees, such as Hoeffding Trees, and it has a monitoring cost of $O(1)$ per instance and sub-linear cost to evaluate split candidates. Previous solutions had a $O(\log n)$ cost per insertion (in the best case) and a linear cost to evaluate split points. Our extensive experimental setup highlights QO's effectiveness in providing accurate split point suggestions while spending much less memory and processing time than its competitors.

An ensemble of learners tends to exceed the predictive performance of individual learners. This approach has been explored for both batch and online learning. Ensembles methods applied to data stream classification were thoroughly investigated over the years, while their regression counterparts received less attention in comparison. In this work, we discuss and analyze several techniques for generating, aggregating, and updating ensembles of regressors for evolving data streams. We investigate the impact of different strategies for inducing diversity into the ensemble by randomizing the input data (resampling, random subspaces and random patches). On top of that, we devote particular attention to techniques that adapt the ensemble model in response to concept drifts, including adaptive window approaches, fixed periodical resets and randomly determined windows. Extensive empirical experiments show that simple techniques can obtain similar predictive performance to sophisticated algorithms that rely on reactive adaptation (i.e., concept drift detection and recovery).

Image segmentation is a key issue in image processing. New image segmentation algorithms have been proposed in the last years. However, there is no optimal algorithm for every image processing task. The selection of the most suitable algorithm usually occurs by testing every possible algorithm or using knowledge from previous problems. These processes can have a high computational cost. Meta-learning has been successfully used in the machine learning research community for the recommendation of the most suitable machine learning algorithm for a new dataset. We believe that meta-learning can also be useful to select the most suitable image segmentation algorithm. This hypothesis is investigated in this paper. For such, we perform experiments with eight segmentation algorithms from two approaches using a segmentation benchmark of 300 images and 2100 augmented images. The experimental results showed that meta-learning can recommend the most suitable segmentation algorithm with more than 80% of accuracy for one group of algorithms and with 69% for the other group, overcoming the baselines used regarding recommendation and segmentation performance.

Transposable Elements (TEs) are DNA sequences capable of moving within a cell's genome. Their transposition has many effects in genomes, such as creating genetic variability and promoting changes in genes' functionality. Recently, TEs classification has been addressed using Machine Learning (ML), more specifically by Hierarchical Classification (HC) methods. Such works proved to be superior than previous ones in the literature. However, there is still room for improvement performance wise. In this direction, Deep Neural Networks (DNNs) have attracted a lot of attention in ML. In particular, Stacked Denoising Auto-Encoders (DAEs) and Deep Multi Layer-Perceptrons (MLPs) are known to provide outstanding results. By performing an extensive evaluation, our results point out that DNNs can enhance the performance of HC methods, being able to push further the state-of-art in TEs' classification.

Hierarchical Classification (HC) consists of classification problems whose classes are structured in a hierarchical fashion. Many problems are addressed by HC, in special a decent amount of works dealt with bioinformatics related problems such as Protein Function Prediction (PFP) and Transposable Elements (TEs) classification. Both of them are still a challenging task for HC due to the noisy and imbalanced nature of the datasets. As a countermeasure, Stacking is an ensemble method capable of generalizing knowledge from many classifiers. In this work, we propose three Stacking methods for HC and evaluate its performance on PFP and TEs datasets. Our results show that, when compared to regular Stacking and state-of-art methods from the literature, our methods are able to obtain superior or competitive performances.