Traditional grammar classifies words based on eight parts of speech: the verb, the noun, the pronoun, the adjective, the adverb, the preposition, the conjunction, and the interjection.
Each part of speech explains not what the word is, but how the word is used. In fact, the same word can be a noun in one sentence and a verb or adjective in the next. The next few examples show how a word's part of speech can change from one sentence to the next, and following them is a series of sections on the individual parts of speech, followed by an exercise.
Books are made of ink, paper, and glue.In this sentence, "books" is a noun, the subject of the sentence.
Deborah waits patiently while Bridget books the tickets.Here "books" is a verb, and its subject is "Bridget."
We walk down the street.In this sentence, "walk" is a verb, and its subject is the pronoun "we."
The mail carrier stood on the walk.In this example, "walk" is a noun, which is part of a prepositional phrase describing where the mail carrier stood.
The town decided to build a new jail.Here "jail" is a noun, which is the object of the infinitive phrase "to build."
The sheriff told us that if we did not leave town immediately he would jail us.Here "jail" is part of the compound verb "would jail."
They heard high pitched cries in the middle of the night.In this sentence, "cries" is a noun acting as the direct object of the verb "heard."
The baby cries all night long and all day long.But here "cries" is a verb that describes the actions of the subject of the sentence, the baby.
The next few sections explain each of the parts of speech in detail. When you have finished, you might want to test yourself by trying the exercise.
Written by Heather MacFadyen
Skip to main content
This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
Learn text moderation concepts
- Article
- 08/09/2022
- 3 minutes to read
In this article
Use Content Moderator's text moderation models to analyze text content, such as chat rooms, discussion boards, chatbots, e-commerce catalogs, and documents.
The service response includes the following information:
- Profanity: term-based matching with built-in list of profane terms in various languages
- Classification: machine-assisted classification into three categories
- Personal data
- Auto-corrected text
- Original text
- Language
Profanity
If the API detects any profane terms in any of the supported languages, those terms are included in the response. The response also contains their location [Index] in the original text. The ListId in the following sample JSON refers to terms found in
custom term lists if available.
"Terms": [
{
"Index": 118,
"OriginalIndex": 118,
"ListId": 0,
"Term": ""
}
Note
For the language parameter, assign eng or leave it empty to see the machine-assisted classification response [preview feature]. This feature supports English only.
For profanity terms detection, use the ISO 639-3 code of the supported languages listed in this article, or leave it empty.
Classification
Content Moderator's machine-assisted text classification feature supports English only, and helps detect potentially undesired content. The flagged content may be assessed as inappropriate depending on context. It conveys the likelihood of each category. The feature uses a trained model to identify possible abusive, derogatory or discriminatory language. This includes slang, abbreviated words, offensive, and intentionally misspelled words.
The following extract in the JSON extract shows an example output:
"Classification": {
"ReviewRecommended": true,
"Category1": {
"Score": 1.5113095059859916E-06
},
"Category2": {
"Score": 0.12747249007225037
},
"Category3": {
"Score": 0.98799997568130493
}
}
Explanation
Category1refers to potential presence of language that may be considered sexually explicit or adult in certain situations.Category2refers to potential presence of language that may be considered sexually suggestive or mature in certain situations.Category3refers to potential presence of language that may be considered offensive in certain situations.Scoreis between 0 and 1. The higher the score, the higher the model is predicting that the category may be applicable. This feature relies on a statistical model rather than manually coded outcomes. We recommend testing with your own content to determine how each category aligns to your requirements.ReviewRecommendedis either true or false depending on the internal score thresholds. Customers should assess whether to use this value or decide on custom thresholds based on their content policies.
Personal data
The personal data feature detects the potential presence of this information:
- Email address
- US mailing address
- IP address
- US phone number
The following example shows a sample response:
"pii":{
"email":[
{
"detected":"",
"sub_type":"Regular",
"text":"",
"index":32
}
],
"ssn":[
],
"ipa":[
{
"sub_type":"IPV4",
"text":"255.255.255.255",
"index":72
}
],
"phone":[
{
"country_code":"US",
"text":"6657789887",
"index":56
}
],
"address":[
{
"text":"1 Microsoft Way, Redmond, WA 98052",
"index":89
}
]
}
Auto-correction
The text moderation response can optionally return the text with basic auto-correction applied.
For example, the following input text has a misspelling.
The quick brown fox jumps over the lazzy dog.
If you specify auto-correction, the response contains the corrected version of the text:
The quick brown fox jumps over the lazy dog.
Creating and managing your custom lists of terms
While the default, global list of terms works great for most cases, you may want to screen against terms that are specific to your business needs. For example, you may want to filter out any competitive brand names from posts by users.
Note
There is a maximum limit of 5 term lists with each list to not exceed 10,000 terms.
The following example shows the matching List ID:
"Terms": [
{
"Index": 118,
"OriginalIndex": 118,
"ListId": 231.
"Term": ""
}
The Content Moderator provides a Term List API with operations for managing custom term lists. Start with the Term Lists API Console and use the REST API code samples. Also check out the Term Lists .NET quickstart if you are familiar with Visual Studio and C#.
Next steps
Test out the APIs with the Text moderation API console.
Feedback
Submit and view feedback for