Classification on 中文AI术语词典https://terms-en.ai-term-hub.com/zh/tags/classification/Recent content in Classification on 中文AI术语词典Hugozh-cnSat, 18 Jul 2026 11:44:45 +0000文本分类https://terms-en.ai-term-hub.com/zh/terms/text_classification/Sat, 18 Jul 2026 11:35:54 +0000https://terms-en.ai-term-hub.com/zh/terms/text_classification/<h2 id="definition">Definition</h2> <p>文本分类是一种监督学习任务，算法为无结构的文本数据分配预定义的类别。常用技术包括朴素贝叶斯、支持向量机和深度学习模型。</p> <h3 id="summary">Summary</h3> <p>根据内容或语义含义将文本归类到不同组织组的过程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>监督学习</li> <li>标注</li> <li>特征提取</li> <li>自然语言处理</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>情感分析</li> <li>垃圾邮件过滤</li> <li>主题建模</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> transformers <span style="color:#f92672">import</span> pipeline </span></span><span style="display:flex;"><span>classifier <span style="color:#f92672">=</span> pipeline(<span style="color:#e6db74">&#34;sentiment-analysis&#34;</span>) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/named-entity-recognition-%E5%91%BD%E5%90%8D%E5%AE%9E%E4%BD%93%E8%AF%86%E5%88%AB/">Named Entity Recognition (命名实体识别)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/sentiment-analysis-%E6%83%85%E6%84%9F%E5%88%86%E6%9E%90/">Sentiment Analysis (情感分析)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/natural-language-processing-%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E5%A4%84%E7%90%86/">Natural Language Processing (自然语言处理)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transformer-models-transformer%E6%A8%A1%E5%9E%8B/">Transformer Models (Transformer模型)</a></li> </ul>序列标注https://terms-en.ai-term-hub.com/zh/terms/sequence_labeling/Sat, 18 Jul 2026 11:33:18 +0000https://terms-en.ai-term-hub.com/zh/terms/sequence_labeling/<h2 id="definition">Definition</h2> <p>序列标注涉及为给定输入序列中的每个标记预测分类标签，例如句子中的单词或字符串中的字符。常见的应用包括词性标注、命名实体识别和句法分块。</p> <h3 id="summary">Summary</h3> <p>一种自然语言处理任务，为输入序列中的每个元素分配一个标签。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>标记分类</li> <li>上下文依赖</li> <li>双向编码</li> <li>条件随机场层</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>命名实体识别</li> <li>词性标注</li> <li>句法分块</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/nlp-%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E5%A4%84%E7%90%86/">NLP (自然语言处理)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transformer/">Transformer</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bilstm-%E5%8F%8C%E5%90%91%E9%95%BF%E7%9F%AD%E6%9C%9F%E8%AE%B0%E5%BF%86%E7%BD%91%E7%BB%9C/">BiLSTM (双向长短期记忆网络)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/crf-%E6%9D%A1%E4%BB%B6%E9%9A%8F%E6%9C%BA%E5%9C%BA/">CRF (条件随机场)</a></li> </ul>线性可分性https://terms-en.ai-term-hub.com/zh/terms/linear_separability/Sat, 18 Jul 2026 11:24:20 +0000https://terms-en.ai-term-hub.com/zh/terms/linear_separability/<h2 id="definition">Definition</h2> <p>线性可分性是指不同类别的数据点可以通过线性边界（如二维空间中的直线或高维空间中的超平面）完全分离的几何条件。</p> <h3 id="summary">Summary</h3> <p>数据集的一种属性，指两个类别可以被一条直线或超平面完美分割。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>超平面</li> <li>决策边界</li> <li>感知机收敛</li> <li>非线性分类</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>评估感知机的适用性</li> <li>核方法选择</li> <li>降维评估</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/support_vector_machine-%E6%94%AF%E6%8C%81%E5%90%91%E9%87%8F%E6%9C%BA/">support_vector_machine (支持向量机)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/kernel_trick-%E6%A0%B8%E6%8A%80%E5%B7%A7/">kernel_trick (核技巧)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/decision_boundary-%E5%86%B3%E7%AD%96%E8%BE%B9%E7%95%8C/">decision_boundary (决策边界)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/perceptron_algorithm-%E6%84%9F%E7%9F%A5%E6%9C%BA%E7%AE%97%E6%B3%95/">perceptron_algorithm (感知机算法)</a></li> </ul>有限记忆AIhttps://terms-en.ai-term-hub.com/zh/terms/limited_memory_ai/Sat, 18 Jul 2026 11:24:07 +0000https://terms-en.ai-term-hub.com/zh/terms/limited_memory_ai/<h2 id="definition">Definition</h2> <p>有限记忆AI代表了AI能力的第二个层次，系统可以从历史数据中学习并相应地调整其行为。与反应式机器不同，这些系统保留有关先前交互的信息，并利用这些信息来改进未来的预测或决策，例如通过监督学习或强化学习算法。</p> <h3 id="summary">Summary</h3> <p>能够存储过去数据和经验以指导未来决策的人工智能系统，构成了大多数现代机器学习应用的基础。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>机器学习</li> <li>数据保留</li> <li>监督学习</li> <li>强化学习</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>垃圾邮件过滤</li> <li>医疗诊断辅助</li> <li>个性化内容推荐</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/reactive_ai-%E5%8F%8D%E5%BA%94%E5%BC%8Fai/">reactive_ai (反应式AI)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/theory_of_mind-%E5%BF%83%E6%99%BA%E7%90%86%E8%AE%BA/">theory_of_mind (心智理论)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/self_aware_ai-%E8%87%AA%E6%88%91%E6%84%8F%E8%AF%86ai/">self_aware_ai (自我意识AI)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/supervised_learning-%E7%9B%91%E7%9D%A3%E5%AD%A6%E4%B9%A0/">supervised_learning (监督学习)</a></li> </ul>基于实例的学习https://terms-en.ai-term-hub.com/zh/terms/instance_based_learning/Sat, 18 Jul 2026 11:22:25 +0000https://terms-en.ai-term-hub.com/zh/terms/instance_based_learning/<h2 id="definition">Definition</h2> <p>也称为基于记忆的学习，该技术不在训练期间构建泛化模型。相反，它存储整个训练数据集。当需要预测时，它会找到最相似的。</p> <h3 id="summary">Summary</h3> <p>一种惰性学习方法，通过将新输入与存储的训练实例进行比较来进行预测。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>惰性学习</li> <li>相似度度量</li> <li>K近邻 (K-Nearest Neighbors)</li> <li>基于记忆</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>推荐系统</li> <li>模式识别</li> <li>中小型数据集</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> sklearn.neighbors <span style="color:#f92672">import</span> KNeighborsClassifier </span></span><span style="display:flex;"><span>knn <span style="color:#f92672">=</span> KNeighborsClassifier(n_neighbors<span style="color:#f92672">=</span><span style="color:#ae81ff">3</span>) </span></span><span style="display:flex;"><span>knn<span style="color:#f92672">.</span>fit(X_train, y_train) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/knn-k-nearest-neighbors/">KNN (K-Nearest Neighbors)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%9B%B8%E4%BC%BC%E5%BA%A6%E6%90%9C%E7%B4%A2-similarity-search/">相似度搜索 (Similarity search)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%83%B0%E6%80%A7%E5%AD%A6%E4%B9%A0-lazy-learning/">惰性学习 (Lazy learning)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%A0%B8%E6%96%B9%E6%B3%95-kernel-methods/">核方法 (Kernel methods)</a></li> </ul>二分类器评估https://terms-en.ai-term-hub.com/zh/terms/evaluation_of_binary_classifiers/Sat, 18 Jul 2026 11:16:37 +0000https://terms-en.ai-term-hub.com/zh/terms/evaluation_of_binary_classifiers/<h2 id="definition">Definition</h2> <p>该领域涉及分析准确率、精确率、召回率、F1分数以及接收者操作特征曲线下面积（AUC-ROC）等指标。它有助于确定模型在区分正负样本方面的表现，并揭示模型在不同阈值下的权衡情况。</p> <h3 id="summary">Summary</h3> <p>评估预测两种可能结果之一的机器学习模型性能的过程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>混淆矩阵</li> <li>精确率-召回率权衡</li> <li>ROC曲线</li> <li>F1分数</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>医学疾病筛查</li> <li>垃圾邮件过滤</li> <li>信用风险评估</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> sklearn.metrics <span style="color:#f92672">import</span> classification_report </span></span><span style="display:flex;"><span>print(classification_report(y_true, y_pred)) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/confusion_matrix-%E6%B7%B7%E6%B7%86%E7%9F%A9%E9%98%B5/">confusion_matrix (混淆矩阵)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/roc_auc-roc%E6%9B%B2%E7%BA%BF%E4%B8%8B%E9%9D%A2%E7%A7%AF/">roc_auc (ROC曲线下面积)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/precision_recall-%E7%B2%BE%E7%A1%AE%E7%8E%87%E4%B8%8E%E5%8F%AC%E5%9B%9E%E7%8E%87/">precision_recall (精确率与召回率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/cross_validation-%E4%BA%A4%E5%8F%89%E9%AA%8C%E8%AF%81/">cross_validation (交叉验证)</a></li> </ul>文档分类https://terms-en.ai-term-hub.com/zh/terms/document_classification/Sat, 18 Jul 2026 11:15:33 +0000https://terms-en.ai-term-hub.com/zh/terms/document_classification/<h2 id="definition">Definition</h2> <p>文档分类是一项基本的自然语言处理任务，算法在此过程中为无结构文本数据分配标签。它涉及从文档中提取特征，并将它们映射到特定的类别中。</p> <h3 id="summary">Summary</h3> <p>根据内容将文本文档归类到预定义组的过程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>文本预处理</li> <li>特征提取</li> <li>监督学习</li> <li>标注</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>电子邮件服务中的垃圾邮件过滤</li> <li>新闻自动标记</li> <li>法律文档分类</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/named-entity-recognition-%E5%91%BD%E5%90%8D%E5%AE%9E%E4%BD%93%E8%AF%86%E5%88%AB/">Named Entity Recognition (命名实体识别)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/text-mining-%E6%96%87%E6%9C%AC%E6%8C%96%E6%8E%98/">Text Mining (文本挖掘)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/natural-language-processing-%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E5%A4%84%E7%90%86/">Natural Language Processing (自然语言处理)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/naive-bayes-%E6%9C%B4%E7%B4%A0%E8%B4%9D%E5%8F%B6%E6%96%AF/">Naive Bayes (朴素贝叶斯)</a></li> </ul>决策列表https://terms-en.ai-term-hub.com/zh/terms/decision_list/Sat, 18 Jul 2026 11:13:51 +0000https://terms-en.ai-term-hub.com/zh/terms/decision_list/<h2 id="definition">Definition</h2> <p>决策列表是一种机器学习模型类型，它将知识表示为条件规则的序列。每个规则由一个条件和预测的类别标签组成。当对新数据进行分类时&hellip;</p> <h3 id="summary">Summary</h3> <p>一种可解释的分类器，由有序规则列表组成，第一个匹配的规则决定输出结果。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>有序规则</li> <li>可解释性</li> <li>首次匹配语义</li> <li>基于规则的分类</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>医疗诊断支持</li> <li>信用评分解释</li> <li>法律合规检查</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%86%B3%E7%AD%96%E6%A0%91-decision-tree/">决策树 (Decision tree)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%9F%BA%E4%BA%8E%E8%A7%84%E5%88%99%E7%9A%84%E7%B3%BB%E7%BB%9F-rule-based-system/">基于规则的系统 (Rule-based system)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%A6%82%E6%9E%9C-%E9%82%A3%E4%B9%88%E8%A7%84%E5%88%99-if-then-rules/">如果-那么规则 (If-then rules)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%8F%AF%E8%A7%A3%E9%87%8A%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0-interpretable-ml/">可解释机器学习 (Interpretable ML)</a></li> </ul>数据集：Yahoo Answers Topicshttps://terms-en.ai-term-hub.com/zh/terms/datasetyahoo_answers_topics/Sat, 18 Jul 2026 11:13:38 +0000https://terms-en.ai-term-hub.com/zh/terms/datasetyahoo_answers_topics/<h2 id="definition">Definition</h2> <p>Yahoo Answers Topics 数据集是更大的雅虎问答档案的子集，专注于组织成不同主题类别的问题和答案。它常用于文本分类、语义相似性分析和非正式语言模式研究。</p> <h3 id="summary">Summary</h3> <p>源自雅虎问答的数据集，包含按特定主题分类的用户生成问题和答案，用于语义匹配和分类任务。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>文本分类</li> <li>语义相似度</li> <li>用户生成内容</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>训练问题分类模型</li> <li>语义文本相似度基准测试</li> <li>分析非正式语言模式</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/ag-news-ag%E6%96%B0%E9%97%BB%E6%95%B0%E6%8D%AE%E9%9B%86/">AG News (AG新闻数据集)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/snli-%E6%96%AF%E5%9D%A6%E7%A6%8F%E8%87%AA%E7%84%B6%E8%AF%AD%E8%A8%80%E6%8E%A8%E7%90%86%E8%AF%AD%E6%96%99%E5%BA%93/">SNLI (斯坦福自然语言推理语料库)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/text-classification-%E6%96%87%E6%9C%AC%E5%88%86%E7%B1%BB/">Text Classification (文本分类)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/semantic-search-%E8%AF%AD%E4%B9%89%E6%90%9C%E7%B4%A2/">Semantic Search (语义搜索)</a></li> </ul>数据集：嵌入数据/QQPhttps://terms-en.ai-term-hub.com/zh/terms/datasetembedding_dataqqp/Sat, 18 Jul 2026 11:12:55 +0000https://terms-en.ai-term-hub.com/zh/terms/datasetembedding_dataqqp/<h2 id="definition">Definition</h2> <p>Quora 问题对（QQP）是一个二元分类数据集，包含来自 Quora 平台的超过 40 万个问题对。其任务是确定两个问题是否具有相同的意图或含义。</p> <h3 id="summary">Summary</h3> <p>Quora 问题对数据集，用于训练模型检测问题之间的语义相似度。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>语义文本相似度</li> <li>二元分类</li> <li>意图匹配</li> <li>句子嵌入</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>重复问题检测</li> <li>微调句子转换器</li> <li>改进 FAQ 机器人</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/sts-benchmark/">STS Benchmark</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%AD%AA%E7%94%9F%E7%BD%91%E7%BB%9C-siamese-networks/">孪生网络 (Siamese Networks)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%96%87%E6%9C%AC%E8%95%B4%E5%90%AB-textual-entailment/">文本蕴含 (Textual Entailment)</a></li> </ul>混淆矩阵https://terms-en.ai-term-hub.com/zh/terms/confusion_matrix/Sat, 18 Jul 2026 11:11:15 +0000https://terms-en.ai-term-hub.com/zh/terms/confusion_matrix/<h2 id="definition">Definition</h2> <p>混淆矩阵是一种特定的表格布局，用于可视化算法（通常是监督学习算法）的性能。它显示了真阳性、真阴性、假阳性和假阴性的计数。</p> <h3 id="summary">Summary</h3> <p>用于描述分类模型在测试数据集上性能的表格。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>真阳性</li> <li>假阴性</li> <li>精确率</li> <li>召回率</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>评估二分类器</li> <li>分析多分类性能</li> <li>调试不平衡数据集中的模型偏差</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">from</span> sklearn.metrics <span style="color:#f92672">import</span> confusion_matrix </span></span><span style="display:flex;"><span>y_true <span style="color:#f92672">=</span> [<span style="color:#ae81ff">2</span>, <span style="color:#ae81ff">0</span>, <span style="color:#ae81ff">2</span>, <span style="color:#ae81ff">2</span>, <span style="color:#ae81ff">0</span>, <span style="color:#ae81ff">1</span>] </span></span><span style="display:flex;"><span>y_pred <span style="color:#f92672">=</span> [<span style="color:#ae81ff">0</span>, <span style="color:#ae81ff">0</span>, <span style="color:#ae81ff">2</span>, <span style="color:#ae81ff">2</span>, <span style="color:#ae81ff">0</span>, <span style="color:#ae81ff">2</span>] </span></span><span style="display:flex;"><span>print(confusion_matrix(y_true, y_pred)) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/precision-%E7%B2%BE%E7%A1%AE%E7%8E%87/">precision (精确率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/recall-%E5%8F%AC%E5%9B%9E%E7%8E%87/">recall (召回率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/f1-score-f1%E5%88%86%E6%95%B0/">F1 score (F1分数)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/roc-curve-roc%E6%9B%B2%E7%BA%BF/">ROC curve (ROC曲线)</a></li> </ul>二分类https://terms-en.ai-term-hub.com/zh/terms/binary_classification/Sat, 18 Jul 2026 11:09:15 +0000https://terms-en.ai-term-hub.com/zh/terms/binary_classification/<h2 id="definition">Definition</h2> <p>二分类是机器学习中一个基本问题，其输出变量是分类变量，恰好有两个可能的结果，例如真/假或垃圾邮件/非垃圾邮件。常用的算法包括逻辑回归、支持向量机等。</p> <h3 id="summary">Summary</h3> <p>一种监督学习任务，目标是为每个输入实例预测两个可能类别之一。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>决策边界</li> <li>混淆矩阵</li> <li>逻辑回归</li> <li>准确率指标</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>垃圾邮件检测</li> <li>医学诊断</li> <li>欺诈检测</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/multiclass-classification-%E5%A4%9A%E5%88%86%E7%B1%BB/">Multiclass Classification (多分类)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/supervised-learning-%E7%9B%91%E7%9D%A3%E5%AD%A6%E4%B9%A0/">Supervised Learning (监督学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/roc-curve-roc%E6%9B%B2%E7%BA%BF/">ROC Curve (ROC曲线)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/precision-%E7%B2%BE%E7%A1%AE%E7%8E%87/">Precision (精确率)</a></li> </ul>Softmaxhttps://terms-en.ai-term-hub.com/zh/terms/softmax/Sat, 18 Jul 2026 11:02:04 +0000https://terms-en.ai-term-hub.com/zh/terms/softmax/<h2 id="definition">Definition</h2> <p>Softmax 广泛用于多分类任务中神经网络的输出层。它接收原始对数几率（logits）向量并对其进行归一化，使每个元素代表一个概率值。</p> <h3 id="summary">Summary</h3> <p>一种数学函数，将任意实数值分数向量转换为概率分布。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>概率分布</li> <li>对数几率 (Logits)</li> <li>归一化</li> <li>多分类</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>图像分类的输出层</li> <li>语言模型的词元预测</li> <li>多标签分类</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span><span style="color:#f92672">import</span> torch </span></span><span style="display:flex;"><span><span style="color:#f92672">import</span> torch.nn.functional <span style="color:#66d9ef">as</span> F </span></span><span style="display:flex;"><span>logits <span style="color:#f92672">=</span> torch<span style="color:#f92672">.</span>tensor([<span style="color:#ae81ff">1.0</span>, <span style="color:#ae81ff">2.0</span>, <span style="color:#ae81ff">3.0</span>]) </span></span><span style="display:flex;"><span>probs <span style="color:#f92672">=</span> F<span style="color:#f92672">.</span>softmax(logits, dim<span style="color:#f92672">=</span><span style="color:#ae81ff">0</span>) </span></span><span style="display:flex;"><span>print(probs) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/argmax-%E5%8F%96%E6%9C%80%E5%A4%A7%E5%80%BC%E7%B4%A2%E5%BC%95/">Argmax (取最大值索引)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E4%BA%A4%E5%8F%89%E7%86%B5%E6%8D%9F%E5%A4%B1-cross-entropy-loss/">交叉熵损失 (Cross-Entropy Loss)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%AF%B9%E6%95%B0%E5%87%A0%E7%8E%87-logits/">对数几率 (Logits)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C-neural-network/">神经网络 (Neural Network)</a></li> </ul>细粒度https://terms-en.ai-term-hub.com/zh/terms/fine_grained/Sat, 18 Jul 2026 10:56:23 +0000https://terms-en.ai-term-hub.com/zh/terms/fine_grained/<h2 id="definition">Definition</h2> <p>细粒度分析涉及在子类别层面而非仅仅在主类别层面识别和分类对象或概念。例如，区分狗的具体品种或不同品牌的智能手机，而不是仅仅将其归类为“动物”或“电子产品”。这要求模型捕捉更细微的特征差异。</p> <h3 id="summary">Summary</h3> <p>描述需要在宽泛类别内区分细微差异的分析或分类任务。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>子分类</li> <li>类内方差</li> <li>详细特征</li> <li>分层标签</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>生态学中的物种鉴定</li> <li>电商中的产品变体检测</li> <li>面部表情的微分析</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/classification-%E5%88%86%E7%B1%BB/">classification (分类)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/visual_features-%E8%A7%86%E8%A7%89%E7%89%B9%E5%BE%81/">visual_features (视觉特征)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/hierarchy-%E5%B1%82%E7%BA%A7%E7%BB%93%E6%9E%84/">hierarchy (层级结构)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/granularity-%E7%B2%92%E5%BA%A6/">granularity (粒度)</a></li> </ul>驱动的https://terms-en.ai-term-hub.com/zh/terms/driven/Sat, 18 Jul 2026 10:50:42 +0000https://terms-en.ai-term-hub.com/zh/terms/driven/<h2 id="definition">Definition</h2> <p>术语“驱动”常作为后缀使用，以指示AI方法背后的主要力量或机制。例如，“数据驱动”意味着决策是基于数据中的统计模式做出的，而“目标驱动”则强调根据特定目标进行优化。</p> <h3 id="summary">Summary</h3> <p>驱动描述了AI系统或方法论，其决策制定或模型架构主要由特定类型的输入（如数据或目标）引导。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>以数据为中心的方法</li> <li>目标优化</li> <li>范式识别</li> <li>强化学习</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>描述以数据为中心的AI工作流程</li> <li>将强化学习代理定义为目标驱动</li> <li>将混合系统分类为知识驱动或数据驱动</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%95%B0%E6%8D%AE%E9%A9%B1%E5%8A%A8-data-driven/">数据驱动 (Data-Driven)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%9B%AE%E6%A0%87%E5%AF%BC%E5%90%91-goal-oriented/">目标导向 (Goal-Oriented)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%9F%BA%E4%BA%8E%E8%A7%84%E5%88%99-rule-based/">基于规则 (Rule-Based)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%90%AF%E5%8F%91%E5%BC%8F-heuristic/">启发式 (Heuristic)</a></li> </ul>