To identify potential customers for various car brands, this paper proposes a comprehensive method based on correlation coefficient and Bayes discriminant to solve related problems. First, this research uses random fo...
详细信息
ISBN:
(数字)9798331527662
ISBN:
(纸本)9798331527679
To identify potential customers for various car brands, this paper proposes a comprehensive method based on correlation coefficient and Bayes discriminant to solve related problems. First, this research uses random forests treatment groups 1 and 2 outliers, based on more than 1000 copies of survey data as the training base for the model; Secondly, the significant analysis is applied for calculating the significant factors for different brands; Next, we need to use the Bayesian model to train the constant term and coefficient vectors. Finally, we need to use the test set data to test the model. The results show that the prediction accuracy for different brands is above 90%.
In recent times, several research works have explored the idea of leveraging machine learning techniques to improve or even replace core components of traditional database architectures, such as the query optimizer an...
详细信息
Exoplanet detection by direct imaging is a difficult task: the faint signals from the objects of interest are buried under a spatially structured nuisance component induced by the host star. The exoplanet signals can ...
详细信息
The research analyzes the needs of efficient data perception and reliable transmission from the perspective of marine environmental monitoring. At the same time, we integrate wireless sensor networks and Mesh networks...
详细信息
ISBN:
(数字)9798331527662
ISBN:
(纸本)9798331527679
The research analyzes the needs of efficient data perception and reliable transmission from the perspective of marine environmental monitoring. At the same time, we integrate wireless sensor networks and Mesh networks, and build multi-sensor heterogeneous wireless ad hoc networks to solve technical problems such as efficient data perception and reliable transmission in complex marine environments. We closely focus on key technical contents such as multi-sensor marine environment detection models, environment-sensitive wireless sensor network MAC protocols, cross-layer and geographical location-assisted routing protocols. The research results have great application value and social benefits for marine economic development and marine environmental protection. The proposed models, protocols and solutions also have strong academic significance for promoting the development of wireless sensor networks.
Traditional recommendation algorithms have problems such as data sparseness and not paying attention to the diversity of recommendation results. In this paper, we use LDA to extract topics of comments about movies, an...
详细信息
Traditional recommendation algorithms have problems such as data sparseness and not paying attention to the diversity of recommendation results. In this paper, we use LDA to extract topics of comments about movies, and identify the emotional tendencies related to topics. As a result, we enrich user interest model and product feature model based on emotional tendencies to improve content-based recommendation algorithms. Most of prior work on applying sentiment classification to recommendation systems only consider the use of sentiment dictionaries to judge polarity, and adopt pattern matching methods to identify features. This paper uses BERT to train sentiment classification models and uses LDA to extract topics. The algorithm is run on the movie review database crawled from Douban, and the experimental result showed that the diversity of recommendation lists had been significantly improved. (C) 2021 The Authors. Published by Elsevier B.V.
New storage requirements, analysis, and visualization of Big data, which includes structured, semi-structured, and unstructured data, have caused the developers in the past decade to begin preferring Big datadatabase...
详细信息
This study investigates the economic impacts of the COVID-19 pandemic on Thailand's tourism-driven economy by analyzing English-language tweets related to tourism from July to December 2020. Utilizing advanced mac...
详细信息
ISBN:
(数字)9798331540364
ISBN:
(纸本)9798331540371
This study investigates the economic impacts of the COVID-19 pandemic on Thailand's tourism-driven economy by analyzing English-language tweets related to tourism from July to December 2020. Utilizing advanced machine learning models-Logistic Regression, Decision Tree Classifier, XGBoost Classifier, and the Pelican Optimization Algorithm-we aimed to understand the sentiments and intentions expressed in these tweets. The research analysis revealed high predictive accuracy, with Logistic Regression achieving 98%, Decision Tree Classifier 99%, XGBoost Classifier 96%, and Pelican Optimization Algorithm 94.5%. These results highlight the importance of addressing concerns related to COVID-19 and political tensions to support the recovery of Thailand's tourism sector. The research also highlights the critical role of information and communications technology (ICT) in managing and revitalizing the tourism industry, especially in the context of reduced consumer spending and travel restrictions. Social media platforms, particularly Twitter, were used to gauge public sentiment and inform recovery strategies. The proposed methodology included extensive data collection, cleaning, and natural language processing, followed by the application of various machine learning algorithms to analyze and interpret sentiment data. The findings offer actionable insights for enhancing tourism marketing strategies and improving the sector's resilience amid ongoing pandemic challenges.
In mobile edge computing (MEC) paradigm, for the security-critical computation tasks offloaded from the mobile device, the MEC server needs to decrypt the encrypted task data before task execution, leading to heavy co...
详细信息
This work contributes with a new approach for tuning hyperparameters of machine learning models, based on sequences of optimization studies based on an initial range of hyperparameters. Through the proposed methodolog...
详细信息
Scenarios dealing with data streams often undergo changes in data distribution, which ultimately lead to a performance degradation of algorithms learning from such data flows (concept drift). This phenomenon calls for...
详细信息
ISBN:
(纸本)9781665424271
Scenarios dealing with data streams often undergo changes in data distribution, which ultimately lead to a performance degradation of algorithms learning from such data flows (concept drift). This phenomenon calls for the adoption of adaptive learning strategies for algorithms to perform resiliently after a change occurs. A multiplicity of approaches have so far addressed this issue by assorted means, e.g. instances weighting, ensembling, instance selection, or parameter tuning, among others. This latter strategy is often neglected as it requires a hyper-parameter tuning process that stream learning scenarios cannot computationally afford in most practical settings. processing times and memory space are usually severely constrained, thus making the tuning phase unfeasible. Consequently, the research community has largely opted for other adaptive strategies with lower computational demands. This work outlines a new perspective to alleviate the hyper-parameter tuning process in the context of concept drift adaptation. We propose two simple and lightweight mechanisms capable of discovering competitive configurations of learning algorithms used for data stream classification. We compare its performance to that of a modern hyper-parametric search method (Successive Halving) over extensive experiments with synthetic and real datasets. We conclude that our proposed methods perform competitively, while consuming less processing time and memory.
暂无评论