A multi-variate time series clustering approach based on intermediate fusion: A case study in air pollution data imputation

Alahamade, Wedad; Lake, Iain; Iglesia, Beatriz de la

doi:10.1016/j.neucom.2021.09.079

Cited by 14 publications

(6 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, a classifier can be constructed on the separable nodes of each layer by the method of layer-bylayer classification, thereby simplifying the model, reducing the difficulty of training, and providing information to assist understanding for manual review.The management of library resources is tasks in library and information work, and manual classification is difficult in the face of massive, diverse and ambiguous books. Therefore technology into the automatic classification of books [21] .…”

Section: Batch Normalized Lstm Modelmentioning

confidence: 99%

Library resource library management based on modern information technology and reconfigurable mobile information systems

Zhang

2023

Preprint

View full text Add to dashboard Cite

Library resource library management based on modern information technology and reconfigurable mobile information systems have attracted much attention from researchers of academia. As a typical application of reconfigurable mobile information systems, the construction of digital library, the introduction of computer automation technology into the field of library resource management, and the realization of automatic classification of book information are of great significance to the realization of digitalization and sharing of book resources. Library resource management is the one and complex tasks in library work. It has been done manually by book authors or book cataloguers for a long time. However, number of books has surged, and the domain knowledge involved in the books has become more and more extensive. It is impossible to manage the library resources manually. In the context of building a digital library, the introduction of computer automation technology into the field of library resource management to realize automatic classification of books and information, and is of great significance to the realization of digitization and sharing of book resources. They have been proven to be advantageous in various fields, including the field of natural language processing. This paper introduces the LSTM model into the field of Chinese library resource management, and tries to propose a classification system with practical application value. The main research work :The Chinese library resource is designed LSTM model, and three experiments are designed on this basis: the coarse-grained book single classification experiment, different measures on the classification results; the fine-grained book single classification experiment, to compare The advantages and disadvantages of direct classification and layer-by-layer classification; coarse-grained book multi-classification experiments to explore the feasibility of multi-classification. After analyzing the experimental results, the model has achieved a good classification effect, and overcomes some deficiencies in the library resource library predecessors, and has a strong practical application value.

show abstract

Section: Batch Normalized Lstm Modelmentioning

confidence: 99%

Library resource library management based on modern information technology and reconfigurable mobile information systems

Zhang

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…After a new sample is obtained, the mode can be directly applied to analyze each sample data, and the processed results are represented by each point set in the vector space. When building a model using neural network, input a set of vectors to all sample nodes [15][16]. Then, according to different situations, we use different methods to extract these information as training vectors to classify and predict the eigenvalues.…”

Section: Data Fusion Technologymentioning

confidence: 99%

Model Construction of Water Pollution Prevention Project Based on Small Sample Learning and Data Fusion

2023

WPPCP

View full text Add to dashboard Cite

At present, the problem of water pollution in China is becoming more and more serious. In view of the scarcity of water resources and deterioration of environmental quality, it is necessary to effectively control pollutants in water bodies. This paper mainly studies the causes of pollutants in small sample data through learning and experiment methods, and takes them as the precondition to connect with the actual environment. On this basis, the improvement measures based on indoor water pollution monitoring and prediction are constructed to analyze, sort out and model the above problems. The emission concentration distribution map obtained by MATLAB software combined with laboratory simulation is used to verify that the above theoretical model is feasible and effective to solve the harmful problems caused by water quality deterioration. The test results show that, based on small sample learning and data fusion technology, It has a certain effect on the water pollution prevention project and can monitor the water pollution.

show abstract

“… xv (12) Then we can obtain , jk v by: (13) For fixed parameter k v , L  is stationary when: (15) From ( 16), we have: (16) Substituting ( 17) into (15), we get:…”

Section: Proposed Fcm Clustering Algorithmmentioning

confidence: 99%

“…As a representative of machine learning based imputation, clustering-based imputation methods have been an increasing research topic nowadays [12,13]. Here the k-means clustering based imputation (KMI) [14] methods are the most general ones.…”

Section: Introductionmentioning

confidence: 99%

Joint clustering and missing value imputation for incomplete data via fuzzy modeling and alternate optimization

Wang

Pang

Zhang

et al. 2023

Preprint

View full text Add to dashboard Cite

Since missing information is an ordinary phenomenon in actual scenarios that increases the difficulty of data analysis, missing value imputation has attracted ever-growing attention in recent years, by exploiting data modeling. Particularly, missing information in engineering design and optimization is a challenging topic. In this work, an exquisite missing value imputation method based on Takagi-Sugeno (TS) fuzzy modeling is proposed, which first divides incomplete dataset by clustering into several fuzzy subsets and finally establishes global model with different regression models on each subset. Specifically, to improve the clustering performance of practical incomplete issues, we incorporate the measurement of common and uncommon observed feature subspace information of data instances into the fuzzy c-means (FCM) framework. Moreover, to better improve the model accuracy, feature selection is introduced to each fuzzy rule and update model parameters and imputations by co-training. The experiments reported both on UCI and tunnel boring machine (TBM) operation datasets demonstrate the proposed method can accurately accomplish missing value imputation of incomplete data.

show abstract

A multi-variate time series clustering approach based on intermediate fusion: A case study in air pollution data imputation

Cited by 14 publications

References 32 publications

Library resource library management based on modern information technology and reconfigurable mobile information systems

Library resource library management based on modern information technology and reconfigurable mobile information systems

Model Construction of Water Pollution Prevention Project Based on Small Sample Learning and Data Fusion

Joint clustering and missing value imputation for incomplete data via fuzzy modeling and alternate optimization

Contact Info

Product

Resources

About