??Kafka-SparkStreamNLP 是一个基于docker容器化管理的实时金融文本分析平台,通过新闻api,采用 Kafka 进行数据流管理,使用 Spark Streaming 结合微调预训练模型finetuning进行NLP处理,并通过输出流将结果存储在clickhouse以便后续使用可视化平台进行olap分析?????????? (2024-04-23, Jupyter Notebook, 0KB, 下载0次)
http://www.pudn.com/Download/item/id/1713827848671067.html
使用不同的DL技术对新闻主题进行分类(NewsSwipe应用程序的扩展)
Categorizing news topics using different DL techniqiues (extension for NewsSwipe app) (2024-04-09, Jupyter Notebook, 0KB, 下载0次)
根据媒体公司《美国新闻》(U.S.News)的报道,探索不同大学特征和大学排名之间的相关性。
Project exploring the correlation between different college features and college rankings according to media company U.S News. (2024-04-09, Jupyter Notebook, 0KB, 下载0次)
配送中心数据库数据分析项目,主要使用SQL和Power BI
Data analysis project about Delivery Center database, mainly using SQL and Power BI (2024-04-08, Jupyter Notebook, 0KB, 下载0次)
使用机器学习和自然语言处理(NLP)检测假新闻。该项目使用Python和NLTK、scikit-learn和WordCloud等库进行文本分析、特征提取和模型训练。探索WordClouds,预处理文本,训练分类器,并评估准确性。
Detect fake news with machine learning and natural language processing (NLP). This project uses Python and libraries like NLTK, scikit-learn, and WordCloud for text analysis, feature extraction, and model training. Explore WordClouds, preprocess text, train classifiers, and evaluate accuracy. (2024-04-09, Jupyter Notebook, 0KB, 下载0次)
Word2Vec黑客新闻
Word2Vec Hackernews (2024-04-08, Jupyter Notebook, 0KB, 下载0次)
使用双向LSTM Word2Vec对假新闻进行分类
Fake News Classify with BidirectionalLSTM Word2Vec (2024-04-08, Jupyter Notebook, 0KB, 下载0次)
零炮推特财经新闻情绪的文本分类
Text Classification on zeroshot twitter financial news sentiment (2024-04-08, Jupyter Notebook, 0KB, 下载0次)
假新闻分类LSTM BiLSTM
Fake News Classify LSTM BiLSTM (2024-04-08, Jupyter Notebook, 0KB, 下载0次)
金融市场新闻情绪分析.ybi
Financial Market News Sentiment Analysis .ybi (2024-04-08, Jupyter Notebook, 0KB, 下载0次)
从头开始创建Word2VecEmbeddings(cbow和skipgram),以预测黑客新闻数据的分数
Creating Word2VecEmbeddings(cbow and skipgram) from scratch to predict scores on hacker news data (2024-04-07, Jupyter Notebook, 0KB, 下载0次)
新闻事件检测
News Event Detection (2024-04-07, Jupyter Notebook, 0KB, 下载0次)
基于新闻向量相似性及多元多元线性回归的故事构建
Story Construction Based on News Vector Similarity and Multiple Linear Regression (2024-04-07, Jupyter Notebook, 0KB, 下载0次)
驭风计划,自然语言处理: 1.Word2Vec TranE的实现(Text8 Wikidata WordSim353); 2.seq2seq模型--机器翻译(NIST); 3.文本情感分析(Rotten Tomato); 4.预训练语言模型实现与应用(DocRED); 5.司法阅读理解(CAIL 2020); 6.面向疫情相关新闻的社会计算应用(CSDC-News); 7.基于 ChatGLM-6B 的本地论文知识库自动问答模型
Wind control plan, natural language processing: 1. Implementation of Word2Vec TranE (Text8 Wikidata WordSim353); 2. seq2seq model -- machine translation (NIST); 3. Text emotion analysis (Rotten Tomato); 4. Implementation and application of pre training language model (DocRED); 5. Judicial Reading Comprehension (CAI 2020); 6. Social computing applications for epidemic related news (CSDC News); 7. Automatic question and answer model of local paper knowledge base based on ChatGLM-6B (2024-04-07, Jupyter Notebook, 0KB, 下载0次)
一个以数据清理和使用Python转换来自Glassdoor的数据科学职位公告为中心的组合项目
A portfolio project centered on data cleaning and transforming data science job postings sourced from Glassdoor using Python (2024-04-06, Jupyter Notebook, 0KB, 下载0次)
该项目的中心是利用社交媒体数据的巨大潜力,在一系列主题中发掘对公众情绪的宝贵见解,特别关注Twitter。在一个社交媒体平台成为公众舆论重要晴雨表的时代。
This project is centered around harnessing the vast potential of social media data to uncover valuable insights into public sentiment across a spectrum of topics, with a special focus on Twitter. In an era where social media platforms serve as significant barometers of public opinion. (2024-04-07, Jupyter Notebook, 0KB, 下载0次)
基于spaCy的假新闻检测模型
Fake News Detection model using spaCy (2024-04-05, Jupyter Notebook, 0KB, 下载0次)
一项研究Python的学术努力,以BBC新闻Web内容作为可选数据集,研究PCA、TSNE和UMAP对PubMed数据集群的影响。它详细研究了降维对K-均值聚类保真度的影响,旨在获得稳健的分析见解。
A scholarly Python endeavor examining PCA, TSNE, UMAP impacts on PubMed data clustering , with BBC News Web Content as optional datasets. It scrutinizes dimensionality reduction s influence on K-means cluster fidelity, aiming for robust analytical insights . (2024-04-06, Jupyter Notebook, 0KB, 下载0次)
该项目使用Hugging Face的NLP框架演示了情感分析和文本生成任务。它展示了文本数据的预处理、使用管道的情感分析以及在新闻文章数据集上使用GPT-2模型的文本生成。它是一个教程,用于使用拥抱面的工具有效地实现NLP任务。
This project demonstrates sentiment analysis and text generation tasks using Hugging Face s NLP framework. It showcases the preprocessing of text data, sentiment analysis with pipelines, and text generation using the GPT-2 model on a dataset of news articles. It serves as a tutorial for implementing NLP tasks efficiently with Hugging Face s tools. (2024-04-06, Jupyter Notebook, 0KB, 下载0次)
NLP模型用于真假新闻的分类和POS标记
NLP model to classify fake news from true news and perform POS Tagging (2024-04-05, Jupyter Notebook, 0KB, 下载0次)