Hybrid neural network

About: Hybrid neural network is a research topic. Over the lifetime, 1305 publications have been published within this topic receiving 18223 citations.

Hybrid neural network and rule-based pattern recognition system capable of self-modification

Abstract: This paper describes a hybrid neural network and rule-based pattern recognition system architecture which is capable of self-modification or learning. The central research issue to be addressed for a multiclassifier hybrid system is whether such a system can perform better than the two classifiers taken by themselves. The hybrid system employs a hierarchical architecture, and it can be interfaced with one or more sensors. Feature extraction routines operating on raw sensor data produce feature vectors which serve as inputs to neural network classifiers at the next level in the hierarchy. These low-level neural networks are trained to provide further discrimination of the sensor data. A set of feature vectors is formed from a concatenation of information from the feature extraction routines and the low-level neural network results. A rule-based classifier system uses this feature set to determine if certain expected environmental states, conditions, or objects are present in the sensors' current data stream. The rule-based system has been given an a priori set of models of the expected environmental states, conditions, or objects which it is expected to identify. The rule-based system forms many candidate directed graphs of various combinations of incoming features vectors, and it uses a suitably chosenmore » metric to measure the similarity between candidate and model directed graphs. The rule-based system must decide if there is a match between one of the candidate graphs and a model graph. If a match is found, then the rule-based system invokes a routine to create and train a new high-level neural network from the appropriate feature vector data to recognize when this model state is present in future sensor data streams. 12 refs., 3 figs.« less

Hybrid neural network text classification method capable of blending abstract with main characteristics

Abstract: The invention relates to a hybrid neural network text classification method capable of blending an abstract with main characteristics. The method comprises the following steps that: step A: extractingan abstract from each text in a training set; step B: using a convolutional neural network to learn the key local features of the abstract obtained in the step A; step C: using a long short-term memory network to learn context time sequence characteristics on the main content of each text in the training set; step D: carrying out cascade connection on two types of characteristics obtained in thestep B and the step C to obtain the integral characteristics of the text, inputting the integral characteristics of each text in the training set into a full connection layer, using a classifier to calculate a probability that each text belongs to each category to train a network, and obtaining a deep neural network model; and step E: utilizing the trained deep neural network model to predict thecategory of a text to be predicted, and outputting the category with a highest probability as a prediction category. The method is favorable for improving text classification accuracy based on the deep neural network.

Signature verification using conic section function neural network

Abstract: This paper presents a new approach for off-line signature verification based on a hybrid neural network (Conic Section Function Neural Network-CSFNN). Artificial Neural Networks (ANNs) have recently become a very important method for classification and verification problems. In this work, CSFNN was proposed for the signature verification and compared with two well known neural network architectures (Multilayer Perceptron-MLP and Radial Basis Function-RBF Networks). The proposed system was trained and tested on a signature database consisting of a total of 304 signature images taken from 8 different persons. A total of 256 samples (32 samples for each person) for training and 48 fake samples (6 fake samples belonging to each person) for testing were used. The results were presented and the comparisons were also made in terms of FAR (False Acceptance Rate) and FRR (False Rejection Rate).

Control of a Batch Polymerization System Using Hybrid Neural Network - First Principle Model

Abstract: In this work, the utilization of neural network in hybrid with first principle models for modelling and control of a batch polymerization process was investigated. Following the steps of the methodology, hybrid neural network (HNN) forward models and HNN inverse model of the process were first developed and then the performance of the model in direct inverse control strategy and internal model control (IMC) strategy was investigated. For comparison purposes, the performance of conventional neural network and PID controller in control was compared with the proposed HNN. The results show that HNN is able to control perfectly for both set points tracking and disturbance rejection studies.

A hybrid model for photovoltaic power prediction of both convolutional and long short-term memory neural networks optimized by genetic algorithm

Abstract: Photovoltaic power generation is affected by weather and geographical environment, showing fluctuations and random multi-interference, and its output power is easy to change with changes in external factors. Therefore, the prediction of output power is crucial to optimize the grid-connected operation of photovoltaic power generation and reduce the impact of uncertainty. This paper proposes a hybrid model of both convolutional neural network (CNN) and long short-term memory neural network (LSTM) based on genetic algorithm (GA) optimization (GA-CNN-LSTM). First, the CNN module is used to extract the spatial features of the data, and then the LSTM module is used to extract the temporal features and nearby hidden states. Optimizing the hyperparameter weights and bias values of the LSTM training network through GA. At the initial stage, the historical data is normalized, and all features were analyzed by grey relational degree. Important features are extracted to reduce the computational complexity of the data. Then, the GA-optimized CNN-LSTM hybrid neural network model (GA-CNN-LSTM) is applied for photovoltaic power prediction experiment. The GA-CNN-LSTM model was compared with the single neural network models such as CNN and LSTM, and the CNN-LSTM hybrid neural network model without GA optimization. Under the Mean Absolute Percentage Error index, the GA-CNN-LSTM algorithm proposed in this paper reduces the error by 1.537% compared with the ordinary single neural network model, and 0.873% compared with the unoptimized CNN-LSTM hybrid neural network algorithm model. From the perspective of training and test running time, the GA-CNN-LSTM model takes a little longer than the single neural network model, but the disadvantage is not obvious. To sum up, the performance of GA-CNN-LSTM model for photovoltaic power predicting is better.