对包含常见词汇和概念的术语表进行语言理解Language understanding glossary of common vocabulary and concepts

语言理解 (LUIS) 术语表解释使用 LUIS 服务时可能遇到的术语。The Language Understanding (LUIS) glossary explains terms that you might encounter as you work with the LUIS service.

活动版本Active version

活动版本是应用的版本,当你使用 LUIS 门户对模型进行更改时,该版本就会更新。The active version is the version of your app that is updated when you make changes to the model using the LUIS portal. 在 LUIS 门户中,如果想对非活动版本进行更改,需先将其设为活动版本。In the LUIS portal, if you want to make changes to a version that is not the active version, you need to first set that version as active.

主动学习Active learning

主动学习是一种机器学习方法,通过该方法,机器学习的模型被用来识别要标记的信息性新示例。Active learning is a technique of machine learning in which the machine learned model is used to identify informative new examples to label. 在 LUIS 中,主动学习是指从当前预测尚不清楚的终结点流量中添加言语,以改进模型。In LUIS, active learning refers to adding utterances from the endpoint traffic whose current predictions are unclear to improve your model. 单击“查看终结点言语”,以查看要标记的言语。Click on "review endpoint utterances", to view utterances to label.

另请参阅:See also:

应用程序(应用)Application (App)

在 LUIS 中,应用程序(或应用)是针对同一数据集构建的机器学习的模型集合,可协同工作以预测特定场景的意向和实体。In LUIS, your application, or app, is a collection of machine learned models, built on the same data set, that works together to predict intents and entities for a particular scenario. 每个应用程序都有一个单独的预测终结点。Each application has a separate prediction endpoint.

如果要构建 HR 机器人,则你可能有一组意向(例如“计划休假时间”、“查询福利”和“更新个人信息”)以及分组到单个应用程序的每个意向的实体。If you are building an HR bot, you might have a set of intents, such as "Schedule leave time", "inquire about benefits" and "update personal information" and entities for each one of those intents that you group into a single application.

创作Authoring

创作是使用 LUIS 门户或创作 API 创建、管理和部署 LUIS 应用的能力。Authoring is the ability to create, manage and deploy a LUIS app, either using the LUIS portal or the authoring APIs.

创作密钥Authoring Key

使用创作密钥创作应用。The authoring key is used to author the app. 不用于生产级别的终结点查询。Not used for production-level endpoint queries. 有关详细信息,请参阅密钥限制For more information, see Key limits.

创作资源Authoring Resource

LUIS 创作资源是可通过 Azure 获取的可管理项。Your LUIS authoring resource is a manageable item that is available through Azure. 资源是你对 Azure 服务的相关创作、训练和发布功能的访问权限。The resource is your access to the associated authoring, training, and publishing abilities of the Azure service. 资源包括访问关联的 Azure 服务所需的身份验证、授权和安全信息。The resource includes authentication, authorization, and security information you need to access the associated Azure service.

创作资源的 Azure 类型为 LUIS-AuthoringThe authoring resource has an Azure "kind" of LUIS-Authoring.

批处理测试Batch test

批处理测试是使用一致且已知的用户话语测试集验证当前 LUIS 应用模型的功能。Batch testing is the ability to validate a current LUIS app's models with a consistent and known test set of user utterances. 批处理测试在 JSON 格式化文件中定义。The batch test is defined in a JSON formatted file.

另请参阅:See also:

F 度量值F-measure

在批处理测试中,它是有关测试准确度的一个度量值。In batch testing, a measure of the test's accuracy.

假负 (FN)False negative (FN)

在批处理测试中,数据点表示某些话语,其中,应用错误地预测了目标意向/实体的缺失。In batch testing, the data points represent utterances in which your app incorrectly predicted the absence of the target intent/entity.

假正 (FP)False positive (FP)

在批处理测试中,数据点表示某些话语,其中,应用错误地预测了目标意向/实体的存在。In batch testing, the data points represent utterances in which your app incorrectly predicted the existence of the target intent/entity.

精度Precision

在批处理测试中,精准率(也称为正预测值)是相关话语在检索到的话语中所占的比例。In batch testing, precision (also called positive predictive value) is the fraction of relevant utterances among the retrieved utterances.

动物批处理测试的一项示例是预测的绵羊数除以动物总数(不管是不是绵羊,都是一样的)。An example for an animal batch test is the number of sheep that were predicted divided by the total number of animals (sheep and non-sheep alike).

RecallRecall

在批处理测试中,召回率(也称为敏感度)是 LUIS 进行通用化的能力。In batch testing, recall (also known as sensitivity), is the ability for LUIS to generalize.

动物批处理测试的一项示例是预测的绵羊数除以可用的绵羊总数。An example for an animal batch test is the number of sheep that were predicted divided by the total number of sheep available.

实报 (TN)True negative (TN)

如果应用正确地预测为无匹配项,则为真负。A true negative is when your app correctly predicts no match. 在批处理测试中,若应用确实针对某个示例预测了一个意图或实体,而该示例却并没有使用该意图或实体进行标记,真负就会出现。In batch testing, a true negative occurs when your app does predict an intent or entity for an example that has not been labeled with that intent or entity.

实报 (TP)True positive (TP)

如果应用正确地预测了匹配项,则为真正。True positive (TP) A true positive is when your app correctly predicts a match. 在批处理测试中,若应用针对某个示例预测了一个意图或实体,且该示例已使用该意图或实体进行了标记,真正就会出现。In batch testing, a true positive occurs when your app predicts an intent or entity for an example that has been labeled with that intent or entity.

分类器Classifier

分类器是一种机器学习的模型,可预测输入适合的类别或类。A classifier is a machine learned model that predicts what category or class an input fits into.

意向是分类器的一个示例。An intent is an example of a classifier.

协作者Collaborator

从概念上讲,协作者与参与者等同。A collaborator is conceptually the same thing as a contributor. 当所有者将协作者的电子邮件地址添加到不受基于角色的访问控制 (RBAC) 的应用中时,该协作者被授予访问权限。A collaborator is granted access when an owner adds the collaborator's email address to an app that isn't controlled with role-based access (RBAC). 如果你仍在使用协作者,则应迁移 LUIS 帐户,并使用 LUIS 创作资源以通过 RBAC 管理参与者。If you are still using collaborators, you should migrate your LUIS account, and use LUIS authoring resources to manage contributors with RBAC.

参与者Contributor

参与者不是应用程序的所有者,但具有相同权限,可以添加、编辑和删除意向、实体和言语。A contributor is not the owner of the app, but has the same permissions to add, edit, and delete the intents, entities, utterances. 参与者为 LUIS 应用提供基于角色的访问 (RBAC)。A contributor provides role-based access (RBAC) to a LUIS app.

另请参阅:See also:

描述符Descriptor

描述符是以前用于机器学习功能的术语。A descriptor is the term formerly used for a machine learning feature.

DomainDomain

在 LUIS 上下文中,域是一个知识领域。In the LUIS context, a domain is an area of knowledge. 你的域特定于你的方案。Your domain is specific to your scenario. 不同的域使用在该域的上下文中具有意义的特定语言和术语。Different domains use specific language and terminology that have meaning in the context of the domain. 例如,如果你正在构建一个用于播放音乐的应用程序,则该应用程序将具有特定于音乐的术语和语言,例如“歌曲、曲目、唱片集、歌词、b 面、艺术家”。For example, if you are building an application to play music, your application would have terms and language specific to music - words like "song, track, album, lyrics, b-side, artist". 有关域的示例,请参阅预生成域For examples of domains, see prebuilt domains.

终结点Endpoint

创作终结点Authoring endpoint

LUIS 创作终结点 URL 是你创作、训练和发布应用的位置。The LUIS authoring endpoint URL is where you author, train, and publish your app. 终结点 URL 包含所发布的应用的区域或自定义子域以及应用 ID。The endpoint URL contains the region or custom subdomain of the published app as well as the app ID.

若要详细了解如何以编程方式创作应用,请参阅开发人员参考Learn more about authoring your app programmatically from the Developer reference

预测终结点Prediction endpoint

LUIS 预测终结点 URL 是在创作并发布 LUIS 应用后提交 LUIS 查询的位置。The LUIS prediction endpoint URL is where you submit LUIS queries after the LUIS app is authored and published. 终结点 URL 包含所发布的应用的区域或自定义子域以及应用 ID。The endpoint URL contains the region or custom subdomain of the published app as well as the app ID. 可以在应用的 Azure 资源页上找到终结点,也可以从获取应用信息 API 获取终结点 URL。You can find the endpoint on the Azure resources page of your app, or you can get the endpoint URL from the Get App Info API.

可以使用 LUIS 预测密钥授权访问预测终结点。Your access to the prediction endpoint is authorized with the LUIS prediction key.

实体Entity

实体是言语中的字词,描述用于实现或识别意向的信息。Entities are words in utterances that describe information used to fulfill or identify an intent. 如果实体很复杂,并且你希望模型标识特定的部分,可将模型分解为子实体。If your entity is complex and you would like your model to identify specific parts, you can break your model into subentities. 例如,你可能希望模型预测某个地址,但同时预测街道、城市、州和邮政编码等子实体。For example, you might want you model to predict an address, but also the subentities of street, city, state, and zipcode. 实体还可以用作模型的功能。Entities can also be used as features to models. 来自 LUIS 应用的响应将包含所预测的意向和所有实体。Your response from the LUIS app will include both the predicted intents and all the entities.

实体提取程序Entity extractor

实体提取程序(有时仅简称为提取器)是 LUIS 用来预测实体的机器学习模型的类型。An entity extractor sometimes known only as an extractor is the type of machine learned model that LUIS uses to predict entities.

实体架构Entity schema

实体架构是你为具有子实体的机器学习实体定义的结构。The entity schema is the structure you define for machine learned entities with subentities. 预测终结点返回在该架构中定义的所有已提取实体和子实体。The prediction endpoint returns all of the extracted entities and subentities defined in the schema.

实体的子实体Entity's subentity

子实体是机器学习实体的子实体。A subentity is a child entity of a machine-learning entity.

非计算机学习实体Non-machine-learning entity

使用文本匹配来提取数据的实体:An entity that uses text matching to extract data:

  • 列表实体List entity
  • 正则表达式实体Regular expression entity

列表实体List entity

列表实体表示一组固定、封闭的相关字词及其同义词。A list entity represents a fixed, closed set of related words along with their synonyms. 与机器学习实体不同,列表实体是精确匹配项。List entities are exact matches, unlike machined learned entities.

如果列表实体中的某个字词包含在列表中,则可预测到该实体。The entity will be predicted if a word in the list entity is included in the list. 例如,如果你有一个名为“大小”的列表实体,并且列表中包含字词“小、中、大”,则无论上下文如何,均可预测使用字词“小”、“中”或“大”的所有言语的大小实体。For example, if you have a list entity called "size" and you have the words "small, medium, large" in the list, then the size entity will be predicted for all utterances where the words "small", "medium", or "large" are used regardless of the context.

正则表达式Regular expression

正则表达式实体表示正则表达式。A regular expression entity represents a regular expression. 与机器学习实体不同,正则表达式实体是精确匹配项。Regular expression entities are exact matches, unlike machined learned entities.

预生成实体Prebuilt entity

请参阅预生成实体的预生成模型条目See Prebuilt model's entry for prebuilt entity

功能Features

在机器学习中,功能是帮助模型识别特定概念的一种特性。In machine learning, a feature is a characteristic that helps the model recognize a particular concept. 这是 LUIS 可以使用的一项提示,但不是硬性规则。It is a hint that LUIS can use, but not a hard rule.

此术语也称为机器学习功能This term is also referred to as a machine-learning feature.

这些提示与标签结合使用,以了解如何预测新数据。These hints are used in conjunction with the labels to learn how to predict new data. LUIS 支持词组列表和使用其他模型作为功能。LUIS supports both phrase lists and using other models as features.

必需功能Required feature

必需功能是一种约束 LUIS 模型输出的方法。A required feature is a way to constrain the output of a LUIS model. 将实体的某项功能被标记为必需时,无论机器学习模型预测什么,该功能都必须存在于要预测的实体示例中。When a feature for an entity is marked as required, the feature must be present in the example for the entity to be predicted, regardless of what the machine learned model predicts.

假设有一个示例,其中你有一个预生成数字功能,且你已将其在点菜机器人的数量实体上标记为必需。Consider an example where you have a prebuilt-number feature that you have marked as required on the quantity entity for a menu ordering bot. 当机器人看到 I want a bajillion large pizzas? 时,不管 bajillion 出现在什么上下文中,都不会被预测为一个数量。When your bot sees I want a bajillion large pizzas?, bajillion will not be predicted as a quantity regardless of the context in which it appears. Bajillion 不是有效的数字,数字预生成实体无法预测到它。Bajillion is not a valid number and won’t be predicted by the number pre-built entity.

IntentIntent

意向表示用户想执行的任务或操作。An intent represents a task or action the user wants to perform. 系指用户输入的内容中所表达的目的或目标,比如预订航班或支付帐单。It is a purpose or goal expressed in a user's input, such as booking a flight, or paying a bill. 在 LUIS 中,整体上将言语分类为意向,但是将部分言语提取为实体In LUIS, an utterance as a whole is classified as an intent, but parts of the utterance are extracted as entities

标签示例Labeling examples

标签或标记是将正示例或负示例与模型关联的过程。Labeling, or marking, is the process of associating a positive or negative example with a model.

意向标签Labeling for intents

在 LUIS 中,应用中的意向是互斥的。In LUIS, intents within an app are mutually exclusive. 这意味着,当你向意向添加言语时,它被认为是该意向的正示例,而对所有其他意向来说它则是负示例 。This means when you add an utterance to an intent, it is considered a positive example for that intent and a negative example for all other intents. 负示例不应与“无”意向混淆,后者表示超出应用范围的言语。Negative examples should not be confused with the "None" intent, which represents utterances that are outside the scope of the app.

实体标签Labeling for entities

在 LUIS 中,你使用一个实体作为正示例来标记意向示例言语中的字词或短语。In LUIS, you label a word or phrase in an intent's example utterance with an entity as a positive example. 标记显示了意向,即应预测该言语的哪些内容。Labeling shows the intent what it should predict for that utterance. 带标记的言语用于训练意向。The labeled utterances are used to train the intent.

LUIS 应用LUIS app

请参阅应用程序(应用)的定义。See the definition for application (app).

模型Model

(机器学习)模型是一种对输入数据进行预测的函数。A (machine learned) model is a function that makes a prediction on input data. 在 LUIS 中,我们将意向分类程序和实体提取程序统称为“模型”,而将经过训练、发布和查询的模型集合统称为“应用”。In LUIS, we refer to intent classifiers and entity extractors generically as "models", and we refer to a collection of models that are trained, published, and queried together as an "app".

规范化值Normalized value

列表实体添加值。You add values to your list entities. 其中每个值都可以包含一个或多个同义词的列表。Each of those values can have a list of one or more synonyms. 仅在响应中返回规范化值。Only the normalized value is returned in the response.

过度拟合Overfitting

若模型只关注特定示例而无法很好地实现通用,就会发生过度拟合。Overfitting happens when the model is fixated on the specific examples and is not able to generalize well.

所有者Owner

每个应用都有一个所有者,即创建应用的人。Each app has one owner who is the person that created the app. 所有者管理 Azure 门户中的应用程序的权限。The owner manages permissions to the application in the Azure portal.

短语列表Phrase list

短语列表是一种特定类型的机器学习功能,包括一组值(字词或短语),它们属于同一个类,并且必须以同样的方式处理它们(例如城市或产品名称)。A phrase list is a specific type of machine learning feature that includes a group of values (words or phrases) that belong to the same class and must be treated similarly (for example, names of cities or products).

预生成模型Prebuilt model

预生成模型是意向、实体或二者的集合,以及带标记的示例。A prebuilt model is an intent, entity, or collection of both, along with labeled examples. 可以将这些常见的预生成模型添加到应用中,以减少应用所需的模型开发工作。These common prebuilt models can be added to your app to reduce the model development work required for your app.

预生成域Prebuilt domain

预生成域是为特定域配置的 LUIS 应用,例如家庭自动化 (HomeAutomation) 或餐厅订位 (RestaurantReservation)。A prebuilt domain is a LUIS app configured for a specific domain such as home automation (HomeAutomation) or restaurant reservations (RestaurantReservation). 已为此域配置意向、话语和实体。The intents, utterances, and entities are configured for this domain.

预生成实体Prebuilt entity

预生成实体是 LUIS 为常用信息类型提供的实体(例如数字、URL 和电子邮件)。A prebuilt entity is an entity LUIS provides for common types of information such as number, URL, and email. 这些是基于公共数据创建的。These are created based on public data. 可以选择将预生成实体添加为独立实体,或添加为实体的功能You can choose to add a prebuilt entity as a stand-alone entity, or as a feature to an entity

预生成意向Prebuilt intent

预生成意向是 LUIS 为常见类型的信息提供的意向,并带有自己的已标记示例言语。A prebuilt intent is an intent LUIS provides for common types of information and come with their own labeled example utterances.

预测Prediction

预测是对 Azure LUIS 预测服务的 REST 请求,它接收新数据(用户言语),并将经过训练和发布的应用程序应用于该数据,以确定找到了哪些意向和实体。A prediction is a REST request to the Azure LUIS prediction service that takes in new data (user utterance), and applies the trained and published application to that data to determine what intents and entities are found.

预测密钥Prediction key

预测密钥(以前称为订阅密钥)是与你在 Azure 中创建的 LUIS 服务相关联的密钥,用于授权你对预测终结点的使用。The prediction key (previously known as the subscription key) is the key associated with the LUIS service you created in Azure that authorizes your usage of the prediction endpoint.

此密钥不是创作密钥。This key is not the authoring key. 如果你有预测终结点密钥,则应为所有终结点请求使用该密钥,而非创作密钥。If you have a prediction endpoint key, it should be used for any endpoint requests instead of the authoring key. 可以在 LUIS 网站的 Azure 资源页底部的终结点 URL 内看到当前的预测密钥。You can see your current prediction key inside the endpoint URL at the bottom of Azure resources page in LUIS website. 它是 subscription-key 名称/值对的值。It is the value of the subscription-key name/value pair.

预测资源Prediction resource

LUIS 预测资源是可通过 Azure 获取的可管理项。Your LUIS prediction resource is a manageable item that is available through Azure. 资源是你对 Azure 服务的相关预测的访问权限。The resource is your access to the associated prediction of the Azure service. 该资源包含预测。The resource includes predictions.

预测资源的 Azure 类型为 LUISThe prediction resource has an Azure "kind" of LUIS.

预测分数Prediction score

分数为介于 0 和 1 之间的数字,这是一个度量值,表示系统对于特定输入言语与特定意向匹配的置信度。The score is a number from 0 and 1 that is a measure of how confident the system is that a particular input utterance matches a particular intent. 接近 1 的分数表示系统对其输出非常有信心,接近 0 的分数表示系统确信输入与特定输出不匹配。A score closer to 1 means the system is very confident about its output and a score closer to 0 means the system is confident that the input does not match a particular output. 分数居中意味着系统对如何做出决定非常不确定。Scores in the middle mean the system is very unsure of how to make the decision.

以一个模型为例,该模型用于标识某些客户文本是否包含食品订单。For example, take a model that is used to identify if some customer text includes a food order. 对于“我想点一杯咖啡”,分数可能为 1(系统非常确信这是一个订单),而对于“我队昨晚赢得了比赛”,分数为 0(系统非常确信这不是一个订单)。It might give a score of 1 for "I'd like to order one coffee" (the system is very confident that this is an order) and a score of 0 for "my team won the game last night" (the system is very confident that this is NOT an order). 对于“我们喝点茶吧”,分数可能为 0.5(不确定这是不是订单)。And it might have a score of 0.5 for "let's have some tea" (isn't sure if this is an order or not).

编程密钥Programmatic key

已重命名为创作密钥Renamed to authoring key.

发布Publish

发布指在暂存或生产终结点上提供一个 LUIS 活动版本。Publishing means making a LUIS active version available on either the staging or production endpoint.

QuotaQuota

LUIS 配额是 Azure 订阅层的限制。LUIS quota is the limitation of the Azure subscription tier. 可同时通过每秒请求数(HTTP 状态 429)和每月请求总数(HTTP 状态 403)来限制 LUIS 配额。The LUIS quota can be limited by both requests per second (HTTP Status 429) and total requests in a month (HTTP Status 403).

架构Schema

架构包括意向和实体以及子实体。Your schema includes your intents and entities along with the subentities. 架构是最初计划好的,之后可随时间的推移进行迭代。The schema is initially planned for then iterated over time. 架构不包括应用设置、功能或示例言语。The schema doesn't include app settings, features, or example utterances.

情绪分析Sentiment Analysis

情绪分析提供文本分析所提供的话语的正值或负值。Sentiment analysis provides positive or negative values of the utterances provided by Text Analytics.

语音启动Speech priming

语音启动使用语音服务改进了对场景中常用的口语和短语的识别。Speech priming improves the recognition of spoken words and phrases that are commonly used in your scenario with Speech Services. 对于启用了语音启动的应用程序,通过为该特定应用程序创建自定义语音模型,可使用所有 LUIS 标记的示例来提高语音识别的准确性。For speech priming enabled applications, all LUIS labeled examples are used to improve speech recognition accuracy by creating a customized speech model for this specific application. 例如,在国际象棋比赛中,你希望确保当用户说“Move knight”时,它不会被解释为“Move night”。For example, in a chess game you want to make sure that when the user says "Move knight", it isn’t interpreted as "Move night". LUIS 应用应包含“knight”被标记为实体的示例。The LUIS app should include examples in which "knight" is labeled as an entity.

初学者密钥Starter key

首次开始使用 LUIS 时要使用的免费密钥。A free key to use when first starting out using LUIS.

同义词Synonyms

在 LUIS 列表实体中,可以创建一个规范化值,每个规范化值可以包含一个同义词列表。In LUIS list entities, you can create a normalized value, which can each have a list of synonyms. 例如,如果你创建一个大小实体,其规范化值有小、中、大和超大。For example, if you create a size entity that has normalized values of small, medium, large, and extra-large. 可以为每个值创建同义词,如下所示:You could create synonyms for each value like this:

规范化值Nomalized value 同义词Synonyms
小型Small 小,8 盎司the little one, 8 ounce
中型Medium 常规,12 盎司regular, 12 ounce
大型Large 大,16 盎司big, 16 ounce
特大Xtra large 最大,24 盎司the biggest one, 24 ounce

当输入中出现任何这些同义词时,该模型将返回实体的相应规范化值。The model will return the normalized value for the entity when any of synonyms are seen in the input.

测试Test

测试 LUIS 应用意味着查看模型预测。Testing a LUIS app means viewing model predictions.

时区偏移Timezone offset

该终结点包含 timezoneOffsetThe endpoint includes timezoneOffset. 这是要从预生成的实体 datetimeV2 删除或向其添加的分钟数。This is the number in minutes you want to add or remove from the datetimeV2 prebuilt entity. 例如,如果话语为“现在几点了?”,则返回的 datetimeV2 是发出客户端请求时的当前时间。For example, if the utterance is "what time is it now?", the datetimeV2 returned is the current time for the client request. 如果客户端请求来自聊天机器人或其他不同于聊天机器人的用户的应用程序,则应传入机器人与该用户之间的时间偏差量。If your client request is coming from a bot or other application that is not the same as your bot's user, you should pass in the offset between the bot and the user.

请参阅更改预生成的 datetimeV2 实体的时区See Change time zone of prebuilt datetimeV2 entity.

标记Token

标记是 LUIS 可识别的最小文本单位。A token is the smallest unit of text that LUIS can recognize. 这在不同语言之间略有不同。This differs slightly across languages.

对于英语,标记是字母和数字的连续跨度(无空格或标点符号)。For English, a token is a continuous span (no spaces or punctuation) of letters and numbers. 空格不是标记。A space is NOT a token.

短语Phrase 令牌计数Token count 说明Explanation
Dog 11 不带标点或空格的单个词。A single word with no punctuation or spaces.
RMT33W 11 记录定位符编号。A record locator number. 它可能包含数字和字母,但没有任何标点。It may have numbers and letters, but does not have any punctuation.
425-555-5555 55 电话号码。A phone number. 每个标点符号都是一个标记,因此 425-555-5555 为 5 个标记:Each punctuation mark is a single token so 425-555-5555 would be 5 tokens:
425
-
555
-
5555
https://luis.ai 77 https
:
/
/
luis
.
ai

定型Train

训练指让 LUIS 了解并习得自上次训练之后对活动版本所做的更改的过程。Training is the process of teaching LUIS about any changes to the active version since the last training.

训练数据Training data

训练数据是训练模型所需的信息集。Training data is the set of information that is needed to train a model. 这包括架构、标记的言语、功能和应用程序设置。This includes the schema, labeled utterances, features, and application settings.

训练错误Training errors

训练错误是指对训练数据的预测与标签不匹配。Training errors are predictions on your training data that do not match their labels.

话语Utterance

言语是用户输入,它是对话中一个句子的简短文本代表。An utterance is user input that is short text representative of a sentence in a conversation. 言语是一条自然语言短语,例如“订 2 张下周二到西雅图的票”。It is a natural language phrase such as "book 2 tickets to Seattle next Tuesday". 通过添加实例言语来训练模型,模型在运行时对新言语进行预测Example utterances are added to train the model and the model predicts on new utterance at runtime

版本Version

LUIS 版本是与 LUIS 应用程序 ID 和已发布终结点关联的 LUIS 应用程序的特定实例。A LUIS version is a specific instance of a LUIS application associated with a LUIS app ID and the published endpoint. 每个 LUIS 应用至少有一个版本。Every LUIS app has at least one version.