Research Article
Spatial Generative Multimodal Dynamic AGI as Model of Consciousness
Evgeny Bryndin*
Issue:
Volume 1, Issue 1, March 2026
Pages:
1-9
Received:
8 November 2025
Accepted:
19 November 2025
Published:
14 February 2026
Abstract: Creating a spatial, generative, multimodal, dynamic artificial general intelligence (AGI) is a complex and multifaceted task. It involves developing a system capable of performing human-like intellectual tasks, such as learning, abstraction, self-regulation, and adaptation. This requires modeling basic cognitive functions and developing methods for long-term storage and retrieval of information. Perception and understanding of the surrounding world: processing sensory data, creating internal representations. Problem solving and planning using solution-finding algorithms. Linguistic thinking and communication: modeling the understanding and generation of natural language. Integration of multilayer and multifaceted models: creating architectures that combine perception, thinking, memory, and motivation. Using a spatial, generative, multimodal, dynamic AGI as a model of consciousness for simulating cognitive processes is planned, using neuromorphic platforms of spiking neural networks and transformers on transformable neurochips. The neuromorphic platform will also facilitate the modeling of metacognitive processes of consciousness, such as the ability to evaluate one's knowledge and strategies. An important stage is, firstly, the creation of test environments to evaluate universality and adaptability; secondly, the gradual increase in task complexity to increase intelligence; and thirdly, the development of infrastructure for large-scale computing. Today, the creation of spatial, generative, multimodal, dynamic artificial general intelligence is considered a feasible, integrative task for multidisciplinary projects. These projects are aimed at, firstly, the ability to solve diverse problems without reprogramming for each specific problem-from data analysis to creative thinking; secondly, the ability to acquire new skills in various ways: independently, through mentoring, and through research; thirdly, maintaining up-to-date information, understanding the situation as a whole, and predicting consequences; fourthly, flexible switching between strategies, choosing the optimal solution under conditions of uncertainty; and fifthly, awareness of one's own cognitive processes, assessing one's knowledge and limitations. The implementation of the projects will require an interdisciplinary international effort of highly qualified scientists, researchers and developers in various fields such as neuroscience, linguistics, artificial intelligence, intelligent modeling and manufacturing based on modern technologies.
Abstract: Creating a spatial, generative, multimodal, dynamic artificial general intelligence (AGI) is a complex and multifaceted task. It involves developing a system capable of performing human-like intellectual tasks, such as learning, abstraction, self-regulation, and adaptation. This requires modeling basic cognitive functions and developing methods for...
Show More
Research Article
Development of a Machine Learning Model That Uses Mine Influents to Soil and Aquarium Water to Predict Future Changes
Kashale Chimanga*,
Christopher Chembe,
Bob Ezekiel Jere
Issue:
Volume 1, Issue 1, March 2026
Pages:
10-18
Received:
22 April 2025
Accepted:
8 May 2025
Published:
14 February 2026
Abstract: The increasing impact of mining activities on aquatic ecosystems has raised serious concerns regarding the accumulation of heavy metals in water bodies, which poses significant risks to fish survival and overall aquaculture sustainability. In regions near mining operations, influents containing metals such as copper (Cu), iron (Fe), and cobalt (Co) can leach into soil and water systems, disrupting water quality. This study was conducted to monitor and predict the physicochemical dynamics of water influenced by mining activities. In-situ measurements of key water quality parameters including pH, Cu, Fe, and Co were carried out using a multi-parameter sensor device in both soil and aquarium water settings to reflect environmental and controlled conditions. The observed concentrations revealed substantial deviations from the optimal levels necessary for healthy aquatic life. To address this, a machine learning (ML) model was developed using the measured influents as input variables to predict future changes in water quality. The predictive model demonstrated high accuracy and potential for real-time application in aquaculture management. Furthermore, linear regression analysis was employed to quantify the relationships between the selected physicochemical parameters and the ideal thresholds for aquatic health, offering deeper insight into their influence on ecosystem stability. The integration of ML for forecasting water quality represents a novel approach to proactive aquaculture monitoring and management, particularly in mining-influenced environments. This research contributes to the growing need for intelligent, data-driven tools in environmental monitoring and supports efforts to mitigate the adverse effects of industrial pollution on aquatic life.
Abstract: The increasing impact of mining activities on aquatic ecosystems has raised serious concerns regarding the accumulation of heavy metals in water bodies, which poses significant risks to fish survival and overall aquaculture sustainability. In regions near mining operations, influents containing metals such as copper (Cu), iron (Fe), and cobalt (Co)...
Show More
Research Article
Anomaly Detection on Numenta Anomaly Benchmark Data Set Using Multiple Machine Learning Algorithms and Impact of Engineered Features on Performance
Abel Channie Demeke*
Issue:
Volume 1, Issue 1, March 2026
Pages:
19-26
Received:
5 December 2025
Accepted:
4 February 2026
Published:
24 February 2026
DOI:
10.11648/j.ajris.20260101.13
Downloads:
Views:
Abstract: This paper evaluates the performance of seven machine learning (ML) algorithms for anomaly detection using the Numenta Anomaly Benchmark (NAB) dataset. The algorithms examined include Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Machines (SVM), Neural Networks (NN), K-Nearest Neighbors (KNN) and Naive Bayes (NB). Two distinct experimental setups were conducted, one evaluating models without additional features and another incorporating created features such as lagged values, rolling window statics, difference values and time based features like hour, day of the year and weekend. The models were trained using the NAB dataset, and their effectiveness in detecting anomalies was assessed. Performance was rigorously evaluated using standard classification metrics like Precision, Recall and F1-Score. In the experiment conducted without additional features, the NN model demonstrated the highest overall performance with an F1-Score of 0.0626526, Precision of 0.0542125 and Recall of 0.0961538 predicting anomalies in 9 files. LR achieved the highest Recall of 0.192029 but with a low Precision of 0.0226541, indicating it often predicted anomalies in a large number of files (38 files) at the cost of high false positives. KNN consistently failed to detect any anomalies across both experiments. The incorporation of additional features generally led to a degradation in performance across most models. For instance, the NN F1-Score decreased to 0.0377358 with features, suggesting that the added features did not enhance and in some cases hindered the models’ anomaly detection capabilities. Some models like LR and SVM also showed an increase in files with errors when features were included. The analysis indicates that while some models are effective at recalling anomalies, they tend to classify a significant amount of normal data as anomaly (low precision). The study highlights the critical impact of feature engineering on anomaly detection performance.
Abstract: This paper evaluates the performance of seven machine learning (ML) algorithms for anomaly detection using the Numenta Anomaly Benchmark (NAB) dataset. The algorithms examined include Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), Support Vector Machines (SVM), Neural Networks (NN), K-Nearest Neighbors (KNN) and Naive Bayes (NB)....
Show More
