[{"data":1,"prerenderedAt":131},["ShallowReactive",2],{"word-biasing":3},{"source":4,"data":5,"quality_score":129,"lastmod":130},"static_word_details",{"lemma":6,"phonetic":7,"pos":9,"definition_zh":11,"usage_points":12,"collocations":34,"examples_basic":80,"examples_advanced":93,"common_mistakes":106,"synonyms_nuance":115,"diff_pairs":124,"word_family":128},"biasing",{"us":8,"uk":8},"",[10],"形容词","在计算机科学中，biasing 指通过调整参数或数据来引入系统性偏差，常用于机器学习或统计模型以优化性能或处理不平衡数据。",[13,20,27],{"title":14,"content":15,"examples":16},"句型搭配","Biasing 通常作为动名词或名词使用，常见于被动语态或与介词搭配，如用于描述模型训练或数据处理中的偏差引入过程。",[17,18,19],"biasing the model","biasing towards a class","biasing against outliers",{"title":21,"content":22,"examples":23},"介词用法","常与介词 towards、against、for 连用，表示偏差的方向或目标，例如 biasing towards positive samples 表示偏向正样本。",[24,25,26],"biasing towards the majority","biasing against noise","biasing for accuracy",{"title":28,"content":29,"examples":30},"语域说明","Biasing 主要出现在学术、技术或专业语境中，特别是计算机科学、数据分析和机器学习领域，日常对话中较少使用。",[31,32,33],"biasing in neural networks","biasing for fairness","biasing in statistical analysis",[35,43,51,59,66,73],{"type":36,"items":37},"verb",[38,39,40,41,42],"apply biasing","introduce biasing","adjust biasing","remove biasing","control biasing",{"type":44,"items":45},"adjective",[46,47,48,49,50],"systematic biasing","intentional biasing","unintended biasing","strong biasing","minimal biasing",{"type":52,"items":53},"noun",[54,55,56,57,58],"data biasing","model biasing","selection biasing","algorithm biasing","bias biasing",{"type":60,"items":61},"adverb",[62,63,64,65],"heavily biasing","slightly biasing","deliberately biasing","inadvertently biasing",{"type":67,"items":68},"preposition",[69,70,71,72],"biasing towards","biasing against","biasing for","biasing in favor of",{"type":74,"items":75},"phrase",[76,77,78,79],"biasing the dataset","biasing the results","biasing the training process","biasing the output",[81,84,87,90],{"en":82,"zh":83},"In this project, we are biasing the data to improve the model's accuracy on rare cases.","在这个项目中，我们正在对数据进行偏置处理，以提高模型在罕见情况下的准确性。",{"en":85,"zh":86},"The algorithm uses biasing to handle the imbalance between positive and negative samples.","该算法使用偏置来处理正负样本之间的不平衡问题。",{"en":88,"zh":89},"You can try biasing the weights to see if it reduces the error rate.","你可以尝试偏置权重，看看是否能降低错误率。",{"en":91,"zh":92},"Biasing the input slightly helped the system perform better in noisy environments.","轻微偏置输入有助于系统在嘈杂环境中表现更好。",[94,97,100,103],{"en":95,"zh":96},"In machine learning, biasing the loss function is a common technique to address class imbalance in classification tasks.","在机器学习中，偏置损失函数是处理分类任务中类别不平衡的常用技术。",{"en":98,"zh":99},"The research paper discusses methods for biasing neural networks to enhance robustness against adversarial attacks.","这篇研究论文讨论了偏置神经网络以增强对抗攻击鲁棒性的方法。",{"en":101,"zh":102},"During the data preprocessing phase, biasing the sampling distribution can mitigate the effects of selection bias in statistical analysis.","在数据预处理阶段，偏置采样分布可以减轻统计分析中选择偏差的影响。",{"en":104,"zh":105},"In business analytics, biasing the model towards recent trends may improve short-term forecasting accuracy but risk overfitting.","在商业分析中，将模型偏置到近期趋势可能提高短期预测准确性，但有过拟合的风险。",[107,111],{"wrong":108,"right":109,"explanation":110},"We need to bias the data for better results.","We need to apply biasing to the data for better results.","错误用法中直接使用 'bias' 作为动词可能不准确，因为 'biasing' 更常作为动名词或名词，表示偏置过程；正确用法强调应用偏置操作。",{"wrong":112,"right":113,"explanation":114},"The biasing is too strong, it causes overfitting.","The biasing is too strong, causing overfitting.","错误用法中使用了 'it causes' 导致句子结构冗余；正确用法使用现在分词 'causing' 使句子更简洁流畅，符合英语表达习惯。",[116,120],{"word":117,"difference":118,"example":119},"skewing","Skewing 更强调数据分布的不对称或倾斜，常用于描述统计偏差；而 biasing 更侧重于人为或系统性引入偏差以达成特定目标，如优化模型。","Skewing the data may distort the analysis, whereas biasing it can be intentional for model tuning.",{"word":121,"difference":122,"example":123},"weighting","Weighting 指分配权重以调整重要性，常用于加权平均或采样；biasing 则更广泛地涉及引入偏差，可能包括权重调整但不限于此，如数据预处理中的偏置。","Weighting the samples gives more importance to certain data points, while biasing might involve altering the entire dataset distribution.",[125,126,127],"bias","skew","weight",[],70,"2026-01-09T06:51:58.489Z",1784307026552]