Alpha-Beta Associative Memories Research Articles

System for Face Recognition under Different Facial Expressions Using a New Associative Hybrid Model Amαβ-KNN for People with Visual Impairment or Prosopagnosia.

Face recognition is a natural skill that a child performs from the first days of life; unfortunately, there are people with visual or neurological problems that prevent the individual from performing the process visually. This work describes a system that integrates Artificial Intelligence which learns the face of the people with whom the user interacts daily. During the study we propose a new hybrid model of Alpha-Beta Associative memories (Amαβ) with Correlation Matrix (CM) and K-Nearest Neighbors (KNN), where the Amαβ-CMKNN was trained with characteristic biometric vectors generated from images of faces from people who present different facial expressions such as happiness, surprise, anger and sadness. To test the performance of the hybrid model, two experiments that differ in the selection of parameters that characterize the face are conducted. The performance of the proposed model was tested in the databases CK+, CAS-PEAL-R1 and Face-MECS (own), which test the Amαβ-CMKNN with faces of subjects of both sexes, different races, facial expressions, poses and environmental conditions. The hybrid model was able to remember 100% of all the faces learned during their training, while in the test in which faces are presented that have variations with respect to those learned the results range from 95.05% in controlled environments and 86.48% in real environments using the proposed integrated system.

Theoretical Foundations for the Alpha-Beta Associative Memories: 10 Years of Derived Extensions, Models, and Applications

The current paper contains the theoretical foundation for the off-the-mainstream model known as Alpha-Beta associative memories ( $$\alpha \beta $$ model). This is an unconventional computation model designed to operate as an associative memory, whose main application is the solution of pattern recognition tasks, particularly for pattern recall and pattern classification. Although this model was devised, proposed and created in 2002, it is worth noting that its theoretical support remains unpublished to this day. This is despite the fact that more than a hundred scientific articles have been published with applications, improvements, and new models derived from the $$\alpha \beta $$ model. The present paper includes all the required definitions, and the rigorous mathematical demonstrations of the lemmas and theorems, explaining the operation of the $$\alpha \beta $$ model, as well as the original models it has inspired or that have been derived from it. Also, brief descriptions of 60 selected articles related to the $$\alpha \beta $$ model are presented. These latter works illustrate the competitiveness (and sometimes superiority) of several extensions and models derived from the original $$\alpha \beta $$ model, when compared against some models and paradigms present in the mainstream current scientific literature.

Pattern classification using smallest normalized difference associative memory

In this paper a new associative classification algorithm is presented. The proposed algorithm overcomes the limitations of the original Alpha-Beta associative memory, while maintaining the fundamental set recalling capacity. This algorithm has two phases. The first phase is based on an Alpha-Beta auto-associative memory, which works in the domain of real numbers, unlike the traditional Alpha-Beta associative memories. In the second phase, normalized difference between the results of first phase and every pattern of the fundamental set is calculated. In order to demonstrate the behaviour and accuracy of the algorithm, multiple well known datasets and classification algorithms have been used. Experimental results have shown that our proposal achieved the best performance in three of the eight pattern classification problems in the medical field, using Stratified 10 Fold cross-validation. Our proposal achieved the best classification accuracy averaged over the all datasets addressed in the present work. Experimental results and statistical significance tests, allow us to affirm that the proposed model is an efficient alternative to perform pattern classification tasks.

Unsupervised Word Sense Disambiguation Using Alpha-Beta Associative Memories

We present an alternative method to the use of overlapping as a distance measure in simple Lesk algorithm. This paper presents an algorithm that uses Alpha-Beta associative memory type Max and Min to measure a given ambiguous word’s meaning in relation to its context, assigning to the word the meaning that is most related. The principal advantage of using this algorithm is the ability to deal with inflectional and derivational forms of words, enabling the possibility of bypassing the stemming procedure of words involved in the disambiguation process. Different experiments were performed, with two parameters as variables: the context window size, and whether stemming was applied or not. The experimental results (F1-score) show that our algorithm performs better than the use of the overlapped metric in the simple Lesk algorithm. Moreover, the experiments show that as more information is added to the sense or meaning, and the overlap metric is used, the precision of the simple Lesk algorithm is decreased-in contrast to the performance of our algorithm.

Data Stream Classification Based on the Gamma Classifier

The ever increasing data generation confronts us with the problem of handling online massive amounts of information. One of the biggest challenges is how to extract valuable information from these massive continuous data streams during single scanning. In a data stream context, data arrive continuously at high speed; therefore the algorithms developed to address this context must be efficient regarding memory and time management and capable of detecting changes over time in the underlying distribution that generated the data. This work describes a novel method for the task of pattern classification over a continuous data stream based on an associative model. The proposed method is based on the Gamma classifier, which is inspired by the Alpha-Beta associative memories, which are both supervised pattern recognition models. The proposed method is capable of handling the space and time constrain inherent to data stream scenarios. The Data Streaming Gamma classifier (DS-Gamma classifier) implements a sliding window approach to provide concept drift detection and a forgetting mechanism. In order to test the classifier, several experiments were performed using different data stream scenarios with real and synthetic data streams. The experimental results show that the method exhibits competitive performance when compared to other state-of-the-art algorithms.

A Novel Text Encryption Algorithm

The encryption/decryption processes is applied to text. Both algorithms involved the use of the associative models, in particular we propose the Alpha-Beta associative memories. The original text is divide in sets of 9 elements, and together with a secret key they build an associative memory which represents the encrypted text. The advantage of our proposal is the cyphertext does not have the same dimension that the plaintext. Additionally, the cyphertext is represented as an image, therefore, since the beginning the encrypted text has a different meaning.

A novel associative model for time series data mining

The paper describes a novel associative model for time series data mining. The model is based on the Gamma classifier, which is inspired on the Alpha–Beta associative memories, which are both supervised pattern recognition models. The objective is to mine known patterns in the time series in order to forecast unknown values, with the distinctive characteristic that said unknown values may be towards the future or the past of known samples. The proposed model performance is tested both on time series forecasting benchmarks and a data set of oil monthly production. Some features of interest in the experimental data sets are spikes, abrupt changes and frequent discontinuities, which considerably decrease the precision of traditional forecasting methods. As experimental results show, this classifier-based predictor exhibits competitive performance. The advantages and limitations of the model, as well as lines of improvement, are discussed.

Time Series Forecasting: Applications to the Upstream Oil and Gas Supply Chain

This paper describes different models which are used for forecasting in the time series context of petroleum engineering. The objective is to reproduce and further predict future oil production in different scenarios in an adjustable time window. Such time series are very similar to those from the sequential manufacturing processes which are usual in many areas of manufacturing industries. We mainly focus on a feedforward neural network model and a Gamma classifier and compare them both on a benchmark and real industrial data under univariate and multivariate settings. While the former model has become recently a standard tool for modeling and prediction, time series forecasting is not the kind of tasks envisioned while designing and developing the Gamma model. The Gamma classifier is inspired on the Alpha-Beta associative memories, taking the alpha and beta operators as basis for the gamma operator. As experimental results show, pattern recognition based classifier shows very competitive performance. The advantages and limitations of each model are discussed.

A New Model of BAM: Alpha-Beta Bidirectional Associative Memories

Most models of Bidirectional associative memories intend to achieve that all trained pattern correspond to stable states; however, this has not been possible. Also, none of the former models has been able to recall all the trained patterns. In this work we introduce a new model of bidirectional associative memory which is not iterative and has no stability problems. It is based on the Alpha-Beta associative memories. This model allows, besides correct recall of noisy patterns, perfect recall of all trained patterns, with no ambiguity and no conditions. An example of fingerprint recognition is presented.

Alpha-Beta Bidirectional Associative Memories

Most models of Bidirectional associative memories intend to achieve that all trained pattern correspond to stable states; however, this has not been possible. Also, none of the former models has been able to recall all the trained patterns. In this work we introduce a new model of bidirectional associative memory which is not iterative and has no stability problems. It is based on the Alpha-Beta associative memories. This model allows perfect recall of all trained patterns, with no ambiguity and no conditions. An example of fingerprint recognition is presented.