Frontiers in Computing and Intelligent Systems ISSN: 2832-6024 | Vol. 5, No. 3, 2023 14 Algorithm Optimization and Performance Improvement of Data Visualization Analysis Platform based on Artificial Intelligence Zepeng Shen Shaanxi University of Technology, Shaanxi 723001, China Abstract: With the rapid development of artificial intelligence, data visualization analysis platforms have been widely applied in various fields. This study mainly explores the optimization and performance improvement of algorithms for data visualization analysis platforms based on artificial intelligence. Firstly, the definition of a data visualization analysis platform and the application of artificial intelligence in it were introduced, and the current problems and challenges were pointed out. Then, a discussion was conducted on algorithm optimization for various stages of research, including data preprocessing, data clustering, data classification, and optimization of data association analysis algorithms. Subsequently, the study proposed some performance improvement methods, including the application of parallel computing technology, distributed computing technology, data compression technology, and data indexing technology. Keywords: Artificial Intelligence; Data Visualization Analysis Platform; Algorithm Optimization; Performance Improvement. 1. Introduction With the arrival of the big data era, data visualization analysis, as an important data analysis method, is widely applied in various fields. Data visualization analysis can transform complex data into intuitive graphics, charts, or visual interfaces, helping users better understand and analyze data, discover hidden patterns and trends in the data, and make corresponding decisions. However, with the continuous increase in data scale and types, traditional data visualization analysis methods are no longer able to meet the demand for large-scale and diverse data. The rapid development of artificial intelligence technology has brought new opportunities and challenges to data visualization analysis. Artificial intelligence can automatically discover patterns and associations in data through technologies such as machine learning, data mining, and pattern recognition, improving the accuracy and efficiency of data visualization analysis [1]. 2. Overview of Artificial Intelligence Data Visualization Analysis Platform 2.1. Definition of Data Visualization Analysis Platform The data visualization analysis platform is a comprehensive platform that integrates artificial intelligence technology and data visualization technology. It aims to transform data into visualized graphics, charts, or interfaces through the processing and analysis of large-scale and diverse data, providing users with intuitive and easy to understand data display and analysis tools. This platform combines artificial intelligence algorithms such as data mining, machine learning, and pattern recognition to automatically discover patterns and associations in data. At the same time, the platform also improves the efficiency and performance of data processing and analysis through technologies such as parallel computing, distributed computing, data compression, and indexing. The goal of a data visualization analysis platform is to provide users with better data understanding and decision support, promoting data-driven decision-making and innovation in various fields [2]. 2.2. Application of Artificial Intelligence in Data Visualization Analysis Platform Artificial intelligence has a wide range of applications in data visualization analysis platforms. Firstly, artificial intelligence technology can help platforms achieve data preprocessing and cleaning, automatically identify and handle anomalies and missing values in data, and improve data accuracy and integrity. Secondly, the application of artificial intelligence in data visualization is mainly reflected in the automatic analysis and visualization of data. Through the automatic learning and data mining of artificial intelligence algorithms, the platform can automatically discover patterns and associations in data, provide targeted data analysis and visual presentation, and help users better understand the data. For example, the platform can automatically identify trends, patterns, and outliers in data, and display them to users through charts and graphs, enabling users to intuitively understand the characteristics and trends of the data. In addition, artificial intelligence can also achieve intelligent recommendation and personalized analysis in data visualization platforms. By analyzing users' historical data and behavior, the platform can provide personalized data analysis and visualization services to users. For example, based on user preferences and needs, the platform can intelligently recommend chart types and presentation methods that are suitable for users, helping them better present and convey data. In addition, artificial intelligence can also achieve automatic report generation and story telling in data visualization platforms. Through automated text generation and storytelling techniques, the platform can automatically generate data reports and stories based on the characteristics and analysis results of the data, helping users better understand and interpret the data [3]. 15 2.3. Current problems and challenges Currently, artificial intelligence data visualization analysis platforms are facing some problems and challenges. The platform still has certain limitations in automatic data processing and cleaning. Although artificial intelligence can help platforms automatically identify and handle anomalies and missing values, it still requires human intervention and optimization in specific fields and complex data situations. Artificial intelligence still has certain errors and limitations in the automatic analysis and visual presentation of data. Due to the complexity and diversity of data, artificial intelligence algorithms may not be able to fully and accurately discover patterns and correlations in the data, resulting in inaccuracies in analysis and visualization results. In addition, artificial intelligence also faces some challenges in intelligent recommendation and personalized analysis in data visualization platforms. Due to the varying needs and preferences of different users, it is a challenge to accurately understand and meet their personalized needs. In addition, artificial intelligence needs to further improve the accuracy and comprehensibility of natural language processing and text generation in terms of automatic report generation and storytelling. The artificial intelligence data visualization analysis platform also faces challenges in data security and privacy protection. Due to the sensitivity and confidentiality of data, ensuring its security and privacy has become an important issue [4]. 3. Algorithm Optimization of Data Visualization Analysis Platform based on Artificial Intelligence 3.1. Optimization of data preprocessing algorithms The algorithm optimization of data visualization analysis platforms based on artificial intelligence mainly focuses on data preprocessing. Data preprocessing is an important part of data analysis, aimed at cleaning and transforming raw data for better subsequent analysis and visualization. When optimizing data preprocessing algorithms, improvements can be made in the following areas[5]. Optimize outlier detection algorithms. Outliers refer to data points that are significantly different from other data points and may interfere with analysis and visualization results. Traditional outlier detection algorithms such as boxplot and Z-score methods can detect some obvious outliers, but their effectiveness is not ideal for complex data distributions and multidimensional data. Therefore, machine learning based outlier detection algorithms such as outlier detection and cluster analysis methods can be introduced to improve the accuracy and robustness of outlier detection. Optimize missing value processing algorithms. In actual data analysis, there are often situations where data is missing, which can lead to inaccuracies in data analysis results. Traditional missing value processing methods such as deletion and interpolation have certain limitations and cannot handle large-scale and complex datasets. Therefore, machine learning based missing value processing algorithms such as random forests and deep learning methods can be introduced to automatically predict and fill in missing values[6]. Optimize data standardization and normalization algorithms. Data may have biases at different scales and ranges, which can have an impact on analysis and visualization results. Traditional data standardization and normalization methods such as Z-score and min-max can handle simple data distributions, but their effectiveness is poor for complex nonlinear data. Therefore, machine learning based standardization and normalization algorithms such as principal component analysis and factor analysis can be introduced to better handle complex data distributions and multidimensional data[7]. 3.2. Optimization of Data Clustering Algorithms In artificial intelligence based data visualization analysis platforms, data classification is a key task. Data classification is the process of assigning data points to predefined categories, which can help users understand and interpret the characteristics and relationships of data. When optimizing data classification algorithms, improvements can be made in the following areas. Optimize the accuracy and robustness of classification algorithms. Traditional classification algorithms such as decision trees and support vector machines suffer from classification errors and overfitting when dealing with complex and noisy data. Therefore, classification algorithms based on deep learning and neural networks, such as convolutional neural networks and recurrent neural networks, can be introduced to improve the accuracy and robustness of classification algorithms. Optimize the interpretability and visualization of classification algorithms. The results generated by traditional classification algorithms are often model weights or scores, which are difficult to understand and interpret. Therefore, classification algorithms based on interpretable machine learning, such as LIME and SHAP, can be introduced to generate classification results with semantic information and achieve more intuitive and interpretable data visualization[8]. Optimize the efficiency and scalability of classification algorithms. Traditional classification algorithms face high computational complexity and memory consumption when dealing with large-scale and high-dimensional data. Therefore, classification algorithms based on approximate and distributed computing, such as approximate nearest neighbors and Spark MLlib, can be introduced to improve the efficiency and scalability of classification algorithms [9]. 3.3. Optimization of Data Classification Algorithms In artificial intelligence based data visualization analysis platforms, data classification is a key task. Data classification is the process of assigning data points to predefined categories, which can help users understand and interpret the characteristics and relationships of data. In order to optimize the data classification algorithm, we can improve it in terms of accuracy, interpretability, efficiency, and interactivity. In order to improve the accuracy and robustness of classification algorithms, classification algorithms based on deep learning and neural networks, such as convolutional neural networks and recurrent neural networks, can be adopted. These algorithms can handle complex and noisy data, improving the accuracy and stability of classification [10]. In order to enhance the interpretability and visualization of classification algorithms, interpretable machine learning classification algorithms such as LIME and SHAP can be introduced. These algorithms can generate classification results with semantic information, enabling users to better 16 understand and explain the reasons and results of classification. At the same time, combined with visualization technology, the classification results can be presented to users in an intuitive manner, improving the visualization effect and user experience. In addition, in order to improve the efficiency and scalability of classification algorithms, approximate and distributed computing methods can be used. Algorithms such as approximate nearest neighbors and Spark MLlib can improve computational efficiency and reduce memory consumption when processing large-scale and high- dimensional data, thereby improving the performance and scalability of classification algorithms. 3.4. Optimization of Data Association Analysis Algorithm Data association analysis is the process of discovering association relationships between data by mining association rules and patterns in a dataset, helping users understand the correlation and potential laws of the data. In order to optimize the data association analysis algorithm, we can improve it in terms of efficiency, accuracy, interpretability, and interactivity. Adopting methods based on sampling and preprocessing. By sampling data, the computational scale can be reduced, thereby improving the algorithm's running speed. At the same time, by preprocessing and optimizing data, such as data compression and indexing techniques, the execution process of algorithms can be accelerated and the efficiency of association rule mining can be improved. Introduce more complex and advanced association analysis algorithms, such as sequence pattern mining and causal association analysis. These algorithms can mine deeper and more complex association rules, thereby improving the accuracy of association analysis and the ability to discover potential patterns. Using interpretable machine learning methods such as rule- based classification and regression. These methods can generate association rules that are easy to understand and interpret, enabling users to better understand the correlation relationships and patterns between data. Introduce the method of interactive correlation analysis. Interactive association analysis enables users to interact more intuitively with association rules, further discovering and verifying association relationships by adjusting parameters and observing results. At the same time, combined with visualization technology, association rules and patterns can be displayed in an intuitive manner, providing a more flexible and personalized data analysis and visualization experience. 4. Performance improvement of data visualization analysis platform based on artificial intelligence 4.1. Application of Parallel Computing Technology The performance improvement of data visualization analysis platform based on artificial intelligence can be achieved by applying parallel computing technology. Parallel computing technology can handle multiple tasks simultaneously, assigning a large number of computing tasks to multiple computing units for parallel processing, thereby significantly improving the efficiency of data analysis and visualization. Utilize parallel computing technology to accelerate data processing. In the data visualization analysis platform, a large amount of data needs to be preprocessed, cleaned, and transformed to meet the visualization requirements. By utilizing parallel computing technology, data can be divided into multiple parts and processed simultaneously on multiple computing units, thereby quickly completing data preprocessing and accelerating data analysis and visualization. Utilize parallel computing technology to accelerate the execution process of algorithms. In data visualization analysis, it is often necessary to apply various complex algorithms to mine the association rules, patterns, and trends of data. The execution process of these algorithms can be very time- consuming. Through parallel computing technology, the computational tasks of the algorithm can be allocated to multiple computing units for parallel processing, thereby quickly completing the execution of the algorithm and improving the efficiency of data visualization analysis. 4.2. Application of Distributed Computing Technology The performance improvement of data visualization and analysis platforms based on artificial intelligence can be achieved by applying distributed computing technology. Distributed computing technology allocates computing tasks to multiple computing nodes, completing tasks through coordination and cooperation to improve computing efficiency and performance. Utilize distributed computing technology to accelerate the processing of large-scale data. In data visualization analysis, there are often challenges in processing massive amounts of data. Through distributed computing technology, data can be divided into multiple parts and allocated to different computing nodes for processing, thereby processing large- scale data in parallel and improving the speed of data analysis and visualization. Utilize distributed computing technology to accelerate the execution of complex algorithms. In data visualization analysis, it is often necessary to apply complex algorithms to mine the association rules and patterns of data. The execution process of these algorithms can be very time-consuming. Through distributed computing technology, the computational tasks of algorithms can be allocated to multiple computing nodes and executed in parallel, thereby accelerating the execution speed of algorithms and improving the efficiency of data visualization analysis. 4.3. Application of Data Compression Technology The data visualization analysis platform based on artificial intelligence can apply data indexing technology in terms of performance improvement. Data indexing technology is an efficient method of data access and query, which can quickly locate and retrieve data by establishing an index structure. In the data visualization analysis platform, the use of data indexing technology can accelerate the speed of data query and analysis, and improve the system's response performance. Data indexing technology can accelerate the process of querying and filtering data. In data visualization analysis platforms, users often need to query and filter data based on specific conditions. By using data indexing technology, index structures can be established to quickly locate eligible data, avoiding the inefficient query method of full table scanning and improving the query speed of data. 17 Data indexing technology can optimize the process of data aggregation and statistical analysis. In data visualization analysis, it is often necessary to aggregate and statistically analyze a large amount of data. By using data indexing technology, an index structure can be established in advance, accelerating the process of data aggregation and statistical analysis, and improving the system's response performance. 4.4. Application of Data Indexing Technology In the performance improvement of data visualization analysis platforms based on artificial intelligence, data compression technology can be applied. Data compression technology is a method of converting data into a more compact form, which can reduce the storage space and transmission bandwidth of data. In the data visualization analysis platform, the use of data compression technology can reduce the storage and transmission costs of data, and improve the performance and efficiency of the system. Data compression technology can reduce the storage space of data. In data visualization analysis platforms, it is usually necessary to process a large amount of data. By using data compression technology, data can be transformed into a more compact form, reducing the storage space requirements of data, thereby saving storage resources and improving data read and write speed. Data compression technology can reduce the transmission bandwidth of data. In data visualization analysis, data transmission is a common bottleneck, especially in poor network environments or large data volumes. By using data compression technology, the volume of data can be reduced, the bandwidth requirements for data transmission can be reduced, and the speed and efficiency of data transmission can be improved. 5. Conclusion The algorithm optimization and performance improvement of data visualization analysis platforms based on artificial intelligence is an important research direction. By applying data compression technology and optimizing data processing algorithms, the efficiency and performance of the system can be improved. In the future, we can further explore the application of data compression technology in data visualization analysis, strengthen algorithm optimization, improve the real-time and scalability of the system, to meet the growing demand for data processing, and promote the application and innovation of artificial intelligence in the field of data visualization analysis. References [1] Wang Meng, Guo Jianlei, Dong Lei. Design and Implementation of a 3D Visualization Analysis Platform for Long March Data [J]. Software, 2023,44 (07): 131-133+156. [2] Tian Xiyu, Wu Shuping. Research on the Construction of Education Big Data Analysis Model and Service Platform - Taking the Teaching Analysis Platform of National Open University as an Example [J]. Journal of Chongqing Open University, 2023,35 (03): 32-41. [3] Wang Hongwei, Yang Kun, Fu Xiang, et al. Design of a Massive Data Mining and Analysis Platform for Fully Mechanized Mining Faces [J]. Industrial and Mining Automation, 2023,49 (05): 30-36+126. DOI: 10.13272/ j.issn.1671-251x.18088. [4] Zhang Ning, Gu Qian, Liang Youcheng, et al. A Mobile Smart Classroom Visualization Platform Based on Big Data Analysis [J]. Information and Computer (Theoretical Edition), 2023,35 (10): 37-39. [5] Li Xiangyu, Li Wenkai, Huo Shitong. Design and development of a hotel information sharing and analysis platform based on data visualization [J]. Software, 2023,44 (04): 68-70. [6] Peng Zigao Design and Implementation of an Access Control and Access Data Analysis Platform Based on Big Data Technology [D]. Southwest University, 2023. DOI: 10.27684/dcnki. gxndx. 2023.001721. [7] Liu Junyang, Zhu Shisong. Design and Application of Station Flight Support Data Visualization Platform [J]. Computer System Application, 2023,32 (05): 67-76. DOI: 10.15888/ j.cnki.csa.009105. [8] Mou Hongyu, Ma Yichong, Huai Xiaoyong. CFD Data Visualization Analysis Service for Cloud Computing Environment [J]. Electronic Technology Application, 2023,49 (03): 118-123. DOI: 10.16157/j.issn.0258-7998.222816. [9] Yin Hang, Liao Ziyuan, Xu Longqin, et al. Design and implementation of a shrimp industry data visualization analysis platform based on ECharts [J]. Modern Agricultural Equipment, 2021,42 (04): 7-14. [10] Cao Xudong Design of a Power Data Visualization Analysis Platform Based on SaaS Mode [D]. Changchun University of Technology, 2019.