Growing Cell Structures Research Articles

Four-dimensional imaging with virtual reality to quantitatively explore jigsaw puzzle-like morphogenesis of Arabidopsis cotyledon pavement cells.

In most dicotyledonous plants, leaf pavement cells exhibit complex jigsaw puzzle-like cell morphogenesis during leaf expansion. Although detailed molecular biological information and mathematical modeling of this jigsaw puzzle-like cell morphogenesis are now available, a full understanding of this process remains elusive. Recent reports have highlighted the importance of three-dimensional (3D) structures (i.e., anticlinal and periclinal cell wall) in understanding the mechanical models that describe this morphogenetic process. We believe that it is important to acquire 3D shapes of pavement cells over time, i.e., acquire and analyze four-dimensional (4D) information when studying the relationship between mechanical modeling and simulations and the actual cell shape. In this report, we have developed a framework to capture and analyze 4D morphological information of Arabidopsis thaliana cotyledon pavement cells by using both direct water immersion observations and computational image analyses, including segmentation, surface modeling, virtual reality and morphometry. The 4D cell models allowed us to perform time-lapse 3D morphometrical analysis, providing detailed quantitative information about changes in cell growth rate and shape, with cellular complexity observed to increase during cell growth. The framework should enable analysis of various phenotypes (e.g., mutants) in greater detail, especially in the 3D deformation of the cotyledon surface, and evaluation of theoretical models that describe pavement cell morphogenesis using computational simulations. Additionally, our accurate and high-throughput acquisition of growing cell structures should be suitable for use in generating in silico model cell structures.

A visualization tool of patent topic evolution using a growing cell structure neural network

This research used a cell structure map to visualize technological evolution and showed the developmental trend in a technological field. The basic concept was to organize patents into a map produced by growing cell structures. The map was then disassembled into clusters with similar contexts using the Girvan---Newman algorithm. Next, the continuity between clusters in two snapshots was identified and used as the base for establishing a trajectory in the technology. An analysis of patents in the flaw detection field found that the field was composed of several technological trajectories. Among them, ultrasonic flaw detection, wafer inspection and substrate inspection were relatively larger and more continuing technologies, while infrared thermography defect inspection has been an emerging topic in recent years. It is to be hoped that the map of technology constructed in this research provides insights into the history of technological evolution and helps explain the transition patterns through changes in cluster continuity. This can serve a reference point by experts who attempt to visualize the mapping of technological development or identify the latest focus of attention.

A novel representation of genomic sequences for taxonomic clustering and visualization by means of self-organizing maps.

Self-organizing maps (SOMs) are readily available bioinformatics methods for clustering and visualizing high-dimensional data, provided that such biological information is previously transformed to fixed-size, metric-based vectors. To increase the usefulness of SOM-based approaches for the analysis of genomic sequence data, novel representation methods are required that automatically and objectively transform aligned nucleotide sequences into numeric vectors, dealing with both nucleotide ambiguity and gaps derived from sequence alignment. Six different codification variants based on Euclidean space, just like SOM processing, have been tested using two SOM models: the classical Kohonen's SOM and growing cell structures. They have been applied to two different sets of sequences: 32 sequences of small sub-unit ribosomal RNA from organisms belonging to the three domains of life, and 44 sequences of the reverse transcriptase region of the pol gene of human immunodeficiency virus type 1 belonging to different groups and sub-types. Our results show that the most important factor affecting the accuracy of sequence clustering is the assignment of an extra weight to the presence of alignment-derived gaps. Although each of the codification variants shows a different level of taxonomic consistency, the results are in agreement with sequence-based phylogenetic reconstructions and anticipate a broad applicability of this codification method.

TOOL CONDITION MONITORING USING ACOUSTIC EMISSION, SURFACE ROUGHNESS AND GROWING CELL STRUCTURES NEURAL NETWORK

The monitoring of tool wear is a most difficult task in the case of various metal-cutting processes. Artificial Neural Networks (ANN) has been used to estimate or classify certain wear parameters, using continuous acquisition of signals from multi-sensor systems. Most of the research has been concentrated on the use of supervised neural network types like multi-layer perceptron (MLP), using back-propagation algorithm and Radial Basis Function (RBF) network. In this article, a new constructive learning algorithm proposed by Fritzke, namely Growing Cell Structures (GCS) has been used for tool wear estimation in face milling operations, thereby monitoring the condition of the tool. GCS generates compact network architecture in less training time and performs well on new untrained data. The performance of this network has been compared with that of another constructive learning algorithm-based neural network, namely the Resource Allocation Network (RAN). For the sake of establishing the effectiveness of GCS, results obtained have been compared with those obtained using Multi Layer Perceptron (MLP), which is a standard and widely used neural network.

Growing Neural Networks using Soft Competitive Learning

ABSTRACT This paper gives an overview of some classical Growing Neural Networks (GNN) using soft competitive learning. In soft competitive learning each input signal is characterized by adapting in addition to the winner also some other neurons of the network. The GNN is also called the ANN with incremental learning. The artificial neural networks (ANN) mapping capability depends on the number of layers and the number of hidden layers in the structure of ANN. There is no formal way of computing network structure. Network structure is usually selected by trial-and-error method but it is time consuming process. Basically, we make use of two mechanisms that may modify the structure of the network: growth and pruning. In this paper, the competitive learning is firstly introduced; secondly the SOM topology and limitations of SOM are illustrated. Thirdly, a class of classical GNN with soft competitive learning is reviewed, such as Neural Gas Network (NGN), Growing Neural Gas (GNG), Self-Organizing Surfaces (SOS), Incremental Grid Growing (lGG), Evolve Self-Organizing Maps (ESOM), Growing Hierarchical Self-Organizing Map (GHSOM), and Growing Cell Structures (GCS).

A combined measure for quantifying and qualifying the topology preservation of growing self-organizing maps

The Self-Organizing Map (SOM) is a neural network model that performs an ordered projection of a high dimensional input space in a low-dimensional topological structure. The process in which such mapping is formed is defined by the SOM algorithm, which is a competitive, unsupervised and nonparametric method, since it does not make any assumption about the input data distribution. The feature maps provided by this algorithm have been successfully applied for vector quantization, clustering and high dimensional data visualization processes. However, the initialization of the network topology and the selection of the SOM training parameters are two difficult tasks caused by the unknown distribution of the input signals. A misconfiguration of these parameters can generate a feature map of low-quality, so it is necessary to have some measure of the degree of adaptation of the SOM network to the input data model. The topology preservation is the most common concept used to implement this measure. Several qualitative and quantitative methods have been proposed for measuring the degree of SOM topology preservation, particularly using Kohonen's model. In this work, two methods for measuring the topology preservation of the Growing Cell Structures (GCSs) model are proposed: the topographic function and the topology preserving map.

Growing fuzzy topology adaptive resonance theory models with a push–pull learning algorithm

A new incrementally growing neural network model, called the growing fuzzy topology ART (GFTART) model, is proposed based on integrating the conventional fuzzy ART model with the incremental topology-preserving mechanism of the growing cell structure (GCS) model. This is in addition, to a new training algorithm, called the push–pull learning algorithm. The proposed GFTART model has two purposes: First, to reduce the proliferation of incrementally generated nodes in the F2 layer by the conventional fuzzy ART model based on replacing each F2 node with a GCS. Second, to enhance the class-dependent clustering representation ability of the GCS model by including the categorization property of the conventional fuzzy ART model. In addition, the proposed push–pull training algorithm enhances the cluster discriminating property and partially improves the forgetting problem of the training algorithm in the GCS model.

P-system communications architecture configuration based on growing self-organizing maps

The different viable architectures that implement P-systems (membrane systems) over distributed clusters of processors have a major drawback: the distribution of these architectures in a balanced tree of processors can minimize external communications and maximize the parallelism grade. For a given P-system and K processors, there exists a great number of possible distributions of membranes. In a recent article, the feasibility of using self-organizing neural networks (SONN) with a growing capacity to help in the selection process of a distribution for a given P-system has been demonstrated, although the nature of the two-dimensional patterns used in the study limited the possibility of defining more flexible degrees of communication, making it more difficult to locate the best distribution. In this article, the capacity of a growing cell structure (GCS) model for projecting high-dimensional spaces in bi-dimensional graphs is explored.

Haptic Perception with Self-Organizing ANNs and an Anthropomorphic Robot Hand

We have implemented and compared four biologically motivated self-organizing haptic systems based on proprioception. All systems employ a 12-d.o.f. anthropomorphic robot hand, the LUCS Haptic Hand 3. The four systems differ in the kind of self-organizing neural network used for clustering. For the mapping of the explored objects, one system uses a Self-Organizing Map (SOM), one uses a Growing Cell Structure (GCS), one uses a Growing Cell Structure with Deletion of Neurons (GCS-DN), and one uses a Growing Grid (GG). The systems were trained and tested with 10 different objects of different sizes from two different shape categories. The generalization abilities of the systems were tested with 6 new objects. The systems showed good performance with the objects from the training set as well as in the generalization experiments. Thus the systems could discriminate individual objects, and they clustered the activities into small cylinders, large cylinders, small blocks, and large blocks. Moreover, the self-organizing ANNs were also organized according to size. The GCS-DN system also evolved disconnected networks representing the different clusters in the input space (small cylinders, large cylinders, small blocks, large blocks), and the generalization samples activated neurons in a proper subnetwork in all but one case.

Construction of Semi-Markov Decision Process Models of Continuous State Space Environments Using Growing Cell Structures and Multiagentk-Certainty Exploration Method

k-certainty exploration method, an efficient reinforcement learning algorithm, is not applied to environments whose state space is continuous because continuous state space must be changed to discrete state space. Our purpose is to construct discrete semi-Markov decision process (SMDP) models of such environments using growing cell structures to autonomously divide continuous state space then usingk-certainty exploration method to construct SMDP models. Multiagentk-certainty exploration method is then used to improve exploration efficiency. Mobile robot simulation demonstrated our proposal's usefulness and efficiency.

A Dynamic and Semantically-Aware Technique for Document Clustering in Biomedical Literature

As an unsupervised learning process, document clustering has been used to improve information retrieval performance by grouping similar documents and to help text mining approaches by providing a high-quality input for them. In this article, the authors propose a novel hybrid clustering technique that incorporates semantic smoothing of document models into a neural network framework. Recently, it has been reported that the semantic smoothing model enhances the retrieval quality in Information Retrieval (IR). Inspired by that, the authors developed and applied a context-sensitive semantic smoothing model to boost accuracy of clustering that is generated by a dynamic growing cell structure algorithm, a variation of the neural network technique. They evaluated the proposed technique on biomedical article sets from MEDLINE, the largest biomedical digital library in the world. Their experimental evaluations show that the proposed algorithm significantly improves the clustering quality over the traditional clustering techniques including k-means and self-organizing map (SOM).

Forecasting the probability of finding oil slicks using a CBR system

A new predicting system is presented in which the aim is to forecast the presence of oil slicks in a certain area of the open sea after an oil spill. Case-based reasoning is a computational methodology designed to generate solutions to a certain problem by analysing previous solutions given to previous solved problems. In this case, the system designed to predict the presence of oil slicks wraps other artificial intelligence techniques such as a radial basis function networks, growing cell structures and principal components analysis in order to develop the different phases of the Case-based reasoning cycle. The proposed system uses information such as sea salinity, sea temperature, wind, currents, pressure, number and area of the slicks,…. obtained from various satellites. The system has been trained using data obtained after the Prestige oil spill, occurred in the Atlantic waters, in the northwest of Spain. The system developed has been able to accurately predict the presence of oil slicks in the north west of the Galician coast, using historical data.

A Biologically Inspired System for Classification of Natural Images

A system for visual information analysis and classification based on a biologically inspired visual selective attention model with knowledge structuring is presented. The system is derived from well-known analogous processes in the visual system of primates and inference procedures of the human brain. It consists of three main units: biologically inspired visual selective attention, knowledge structuring, and clustering of visual information. The biologically inspired visual selective attention unit closely follow the mechanisms of the visual what pathway and where pathway in the primates' brain. It uses a bottom-up approach to generate a salient area based on low-level features extracted from natural images. The scale selection to determine suitable size of salient areas uses a maximum entropy approach. This unit also contains a low-level top-down selective attention module that performs decisions on interesting objects by human interaction. In this module, a reinforcement/inhibition mechanism is exploited. The knowledge structuring unit automatically creates a relevance map from salient image areas generated by the biologically inspired unit. It also derives a set of well-structured representations from low-level descriptions to drive the final classification. The knowledge structuring unit relys on human knowledge to produce suitable links between low-level descriptions and high-level representation on a limited training set. The backbone of this unit is a distribution mapping strategy involving two basic modules: structured low-level feature extraction using convolution neural network and a topology representation module based on a growing cell structure network. The third unit of the system classification is achieved by simulating high-level top-down visual information perception and clustering using an incremental Bayesian parameter estimation method. The proposed modular system architecture offers straightforward expansion to include user relevance feedback, contextual input, and multimodal information if available

Visualization of Volcanic Rock Geochemical Data and Classification with Artificial Neural Networks

An unsupervised neural network technique, Growing Cell Structures (GCS) was used to visualize geochemical differences between four different island arc volcanic rock types: basalts, andesites, dacites and rhyolites. The output of the method shows the cluster structure of the dataset clearly, and the relevant geochemical patterns and relationships between its variables. The data can be separated into four clusters, each associated with a specific volcanic rock type (basalt, andesite, dacite and rhyolite), according to a unique combination of major element concentrations. Following clustering, performance of the trained GCS network as a classifier of volcanic rock type was tested using two test datasets with major element concentration data for 312 and 496 island arc volcanic rock samples of known volcanic type. Preliminary classification results are promising. In the first test dataset 94% of basalts, 76% of andesites, 83% of dacites and 100% of the rhyolites were classified correctly. Successful classification rates in the second dataset were 100%, 80%, 77%, and 98% respectively. The success of the analysis suggests that neural networks analysis constitutes a useful analytical tool for identification of natural clusters and examination of the relationships between numeric variables in large datasets, and that can be used for automatic classification of new data.

Gene‐CBR: A CASE‐BASED REASONIG TOOL FOR CANCER DIAGNOSIS USING MICROARRAY DATA SETS

Gene expression profiles are composed of thousands of genes at the same time, representing the complex relationships between them. One of the well‐known constraints specifically related to microarray data is the large number of genes in comparison with the small number of available experiments or cases. In this context, the ability of design methods capable of overcoming current limitations of state‐of‐the‐art algorithms is crucial to the development of successful applications. This paper presents gene‐CBR, a hybrid model that can perform cancer classification based on microarray data. The system employs a case‐based reasoning model that incorporates a set of fuzzy prototypes, a growing cell structure network and a set of rules to provide an accurate diagnosis. The hybrid model has been implemented and tested with microarray data belonging to bone marrow cases from forty‐three adult patients with cancer plus a group of six cases corresponding to healthy persons.

An adaptive signal compression system with pre-specified reconstruction quality and compression rate

Two essential properties of a signal compression method are the compression rate and the distance between the original signal and the reconstruction from the compressed signal. These two properties are used to assess the performance and quality of the method. In a recent work [B. Tümer, B. Demiröz, Lecture Notes in Computer Science-Computer and Information Sciences, volume 2869, chapter Signal Compression Using Growing Cell Structures: A Transformational Approach, Springer Verlag, 2003, pp. 952–959], an adaptive signal compression system (ACS) is presented which defines the performance of the system as a function of the system complexity, system sensitivity and data size. For a compression method, it is desirable to formulate the performance of the system as a function of the system complexity and sensitivity to optimize the performance of the system. It would be further desirable to express the reconstruction quality in terms of the same system parameters so as to know up front what compression rate to end up with for a specific reconstruction quality. In this work, we modify ACS such that the modified ACS (MACS) estimates the reconstruction quality for a given system complexity and sensitivity. Once this relation is identified it is possible to optimize either compression rate or reconstruction quality with respect to system sensitivity and system complexity while limiting the other one.

Modeling user multiple interests by an improved GCS approach

User interest profile is the crucial component of most personalized recommender systems. The diversity and time-dependent evolving nature of user interests are creating difficulties in constructing and maintaining a sound user profile. This paper presents a simple but effective model, by using improved growing cell structures (GCS), to address this problem. The GCS is a kind of self-organizing map neural network with changeable network structure. By virtue of the clustering and structure adaptation capability of GCS, the proposed model maps the problem of learning and keeping track of user interests into a clustering and cluster-maintaining problem. Each cluster found by GCS represents an interest category of a user and the cluster maintaining, including cluster addition and deletion, corresponds to the addition of user's new interests and the removal of user's old interests. The proposed model has been validated by a set of experiments performed on a benchmark dataset. Results from experiments show that our model provides reasonable performance and high adaptability for learning user multiple interests and their changes.

Deterministic projection by growing cell structure networks for visualization of high-dimensionality datasets

Recent advances in clinical proteomics data acquisition have led to the generation of datasets of high complexity and dimensionality. We present here a visualization method for high-dimensionality datasets that makes use of neuronal vectors of a trained growing cell structure (GCS) network for the projection of data points onto two dimensions. The use of a GCS network enables the generation of the projection matrix deterministically rather than randomly as in random projection. Three datasets were used to benchmark the performance and to demonstrate the use of this deterministic projection approach in real-life scientific applications. Comparisons are made to an existing self-organizing map projection method and random projection. The results suggest that deterministic projection outperforms existing methods and is suitable for the visualization of datasets of very high dimensionality.

The Evolving Tree?A Novel Self-Organizing Network for Data Analysis

The Self-Organizing Map (SOM) is one of the best known and most popular neural network-based data analysis tools. Many variants of the SOM have been proposed, like the Neural Gas by Martinetz and Schulten, the Growing Cell Structures by Fritzke, and the Tree-Structured SOM by Koikkalainen and Oja. The purpose of such variants is either to make a more flexible topology, suitable for complex data analysis problems or to reduce the computational requirements of the SOM, especially the time-consuming search for the best-matching unit in large maps. We propose here a new variant called the Evolving Tree which tries to combine both of these advantages. The nodes are arranged in a tree topology that is allowed to grow when any given branch receives a lot of hits from the training vectors. The search for the best matching unit and its neighbors is conducted along the tree and is therefore very efficient. A comparison experiment with high dimensional real world data shows that the performance of the proposed method is better than some classical variants of SOM.

FSfRT: Forecasting System for Red Tides

A hybrid neuro-symbolic problem-solving model is presented in which the aim is to forecast parameters of a complex and dynamic environment in an unsupervised way. In situations in which the rules that determine a system are unknown, the prediction of the parameter values that determine the characteristic behaviour of the system can be a problematic task. In such a situation, it has been found that a hybrid case-based reasoning system can provide a more effective means of performing such predictions than other connectionist or symbolic techniques. The system employs a case-based reasoning model to wrap a growing cell structures network, a radial basis function network and a set of Sugeno fuzzy models to provide an accurate prediction. Each of these techniques is used at a different stage of the reasoning cycle of the case-based reasoning system to retrieve historical data, to adapt it to the present problem and to review the proposed solution. This system has been used to predict the red tides that appear in the coastal waters of the north west of the Iberian Peninsula. The results obtained from experiments, in which the system operated in a real environment, are presented.