sharepoint syntex
TRANSCRIPT
SharePoint Syntex Page 1
Contoso Corporation uses SharePoint to store a huge variety of documents, from sales and marketing to HR and Operations. Now they’re implementing SharePoint Syntex to automatically capture knowledge from business documents and use that to drive understanding and enrich automation across the company.
In this demo, Megan Bowen, an HR Manager at Contoso, will train SharePoint Syntex to recognize a common Contoso HR document, automatically extract useful information from it and store that data.
Page 2
Megan Bowen, an HR manager at Contoso, is going to create a document understanding model that will process a typical HR document and extract key terms.
Page 3
Page 4
Page 5
Today, Megan is going to train the AI to recognize insurance benefit change letters. This is a common document that Contoso is required to send when employees’ "insurance benefits change". She makes this a new content type in SharePoint. She could also create a model linked to an existing content type.
Page 6
Megan can also configure the model to automatically apply the correct document retention policies whenever a document of this type is identified. This will greatly simplify compliance for Megan’s department.
Page 7
Now Megan creates the model. She can also use Syntex to extract and process structured content, such as forms.
Page 8
Syntex uses AI, trained on existing documents in order to classify and recognize similar documents in the future.
Page 9
To begin this process, Megan uploads several sample documents for training.
Page 10
Page 11
Page 12
The training documents include several examples of benefit change letters, as well as some other HR documents that aren’t benefit letters to provide contrast.
Page 13
Page 14
Page 15
Megan now creates a classifier to identify if an entire document belongs or does not belong to a specific content type.
Page 16
Page 17
She goes through each document in the training set and marks all the ones that are benefit change letters.
Page 18
Page 19
Page 20
Page 21
The last two documents are not benefit letters. Megan marks them as such.
Page 22
Page 23
Finally, she adds an explanation to help define information in Syntex.
Page 24
Page 25
Page 26
Page 27
Page 28
Page 29
Here, Megan uses the phrase Benefit Change Notice, which appears in all the benefit letters, to help identify them.
Page 30
Page 31
When she’s done, Megan can see that the classifier has processed each training document and training is complete.
Page 32
Now Megan will create an extractor to identify and pull a specific piece of information from the document—in this case, the insurance provider.
Page 33
The extractor will save this information to a new column in a SharePoint library, for easier discovery and use.
Page 34
Page 35
Page 36
Megan trains the AI to recognize the insurance provider name. In each document, she highlights the information she wants to extract.
Page 37
Page 38
Page 39
Page 40
Page 41
Page 42
Page 43
Page 44
Page 45
Page 46
Page 47
Page 48
The last two documents are negative examples and do not include the information that Megan wants to extract.
Page 49
Page 50
Page 51
Page 52
Page 53
Page 54
Page 55
Page 56
Page 57
Page 58
Page 59
Page 60
Page 61
When she’s done, Megan reviews the results to see how the extractor preforms. It catches most of the providers, but she can see the extractor didn’t fully capture one of the providers, Best for you Organics.
Page 62
Page 63
Page 66
Page 65
Page 66
Page 67
Page 68
Page 69
Page 70
With both the classifier and extractor working as desired, Megan is ready to apply this model to a library.
Page 71
When applied to a library, the model will automatically process every document added to the library.
Page 72
If the document fits the classification, the relevant information is extracted and stored in a SharePoint column.
Page 73
Page 74
Page 75
Page 76
To test her model, Megan uploads some more benefit change letters.
Page 77
Page 78
Page 79 End
After a few moments, Syntex has successfully identified them as benefit change letters and extracted the insurance provider for each one.
SharePoint Syntex helps organizations like Contoso transform content into knowledge and automate content processing using advanced AI and machine learning combined with human experience.
SharePoint Syntex
Page 1
Page 2
Page 3
Page 4
Page 5
Page 6
Page 7
Page 8
Page 9
Page 10
Page 11
Page 12
Page 13
Page 14
Page 15
Page 16
Page 17
Page 18
Page 19
Page 20
Page 21
Page 22
Page 23
Page 24
Page 25
Page 26
Page 27
Page 28
Page 29
Page 30
Page 31
Page 32
Page 33
Page 34
Page 35
Page 36
Page 37
Page 38
Page 39
Page 40
Page 41
Page 42
Page 43
Page 44
Page 45
Page 46
Page 47
Page 48
Page 49
Page 50
Page 51
Page 52
Page 53
Page 54
Page 55
Page 56
Page 57
Page 58
Page 59
Page 60
Page 61
Page 62
Page 63
Page 66
Page 65
Page 66
Page 67
Page 68
Page 69
Page 70
Page 71
Page 72
Page 73
Page 74
Page 75
Page 76
Page 77
Page 78
Contoso Corporation uses SharePoint to store a huge variety of documents, from sales and marketing to HR and Operations. Now they’re implementing SharePoint Syntex to automatically capture knowledge from business documents and use that to drive understanding and enrich automation across the company.
In this demo, Megan Bowen, an HR Manager at Contoso, will train SharePoint Syntex to recognize a common Contoso HR document, automatically extract useful information from it and store that data.
Page 2
Megan Bowen, an HR manager at Contoso, is going to create a document understanding model that will process a typical HR document and extract key terms.
Page 3
Page 4
Page 5
Today, Megan is going to train the AI to recognize insurance benefit change letters. This is a common document that Contoso is required to send when employees’ "insurance benefits change". She makes this a new content type in SharePoint. She could also create a model linked to an existing content type.
Page 6
Megan can also configure the model to automatically apply the correct document retention policies whenever a document of this type is identified. This will greatly simplify compliance for Megan’s department.
Page 7
Now Megan creates the model. She can also use Syntex to extract and process structured content, such as forms.
Page 8
Syntex uses AI, trained on existing documents in order to classify and recognize similar documents in the future.
Page 9
To begin this process, Megan uploads several sample documents for training.
Page 10
Page 11
Page 12
The training documents include several examples of benefit change letters, as well as some other HR documents that aren’t benefit letters to provide contrast.
Page 13
Page 14
Page 15
Megan now creates a classifier to identify if an entire document belongs or does not belong to a specific content type.
Page 16
Page 17
She goes through each document in the training set and marks all the ones that are benefit change letters.
Page 18
Page 19
Page 20
Page 21
The last two documents are not benefit letters. Megan marks them as such.
Page 22
Page 23
Finally, she adds an explanation to help define information in Syntex.
Page 24
Page 25
Page 26
Page 27
Page 28
Page 29
Here, Megan uses the phrase Benefit Change Notice, which appears in all the benefit letters, to help identify them.
Page 30
Page 31
When she’s done, Megan can see that the classifier has processed each training document and training is complete.
Page 32
Now Megan will create an extractor to identify and pull a specific piece of information from the document—in this case, the insurance provider.
Page 33
The extractor will save this information to a new column in a SharePoint library, for easier discovery and use.
Page 34
Page 35
Page 36
Megan trains the AI to recognize the insurance provider name. In each document, she highlights the information she wants to extract.
Page 37
Page 38
Page 39
Page 40
Page 41
Page 42
Page 43
Page 44
Page 45
Page 46
Page 47
Page 48
The last two documents are negative examples and do not include the information that Megan wants to extract.
Page 49
Page 50
Page 51
Page 52
Page 53
Page 54
Page 55
Page 56
Page 57
Page 58
Page 59
Page 60
Page 61
When she’s done, Megan reviews the results to see how the extractor preforms. It catches most of the providers, but she can see the extractor didn’t fully capture one of the providers, Best for you Organics.
Page 62
Page 63
Page 66
Page 65
Page 66
Page 67
Page 68
Page 69
Page 70
With both the classifier and extractor working as desired, Megan is ready to apply this model to a library.
Page 71
When applied to a library, the model will automatically process every document added to the library.
Page 72
If the document fits the classification, the relevant information is extracted and stored in a SharePoint column.
Page 73
Page 74
Page 75
Page 76
To test her model, Megan uploads some more benefit change letters.
Page 77
Page 78
Page 79 End
After a few moments, Syntex has successfully identified them as benefit change letters and extracted the insurance provider for each one.
SharePoint Syntex helps organizations like Contoso transform content into knowledge and automate content processing using advanced AI and machine learning combined with human experience.
SharePoint Syntex
Page 1
Page 2
Page 3
Page 4
Page 5
Page 6
Page 7
Page 8
Page 9
Page 10
Page 11
Page 12
Page 13
Page 14
Page 15
Page 16
Page 17
Page 18
Page 19
Page 20
Page 21
Page 22
Page 23
Page 24
Page 25
Page 26
Page 27
Page 28
Page 29
Page 30
Page 31
Page 32
Page 33
Page 34
Page 35
Page 36
Page 37
Page 38
Page 39
Page 40
Page 41
Page 42
Page 43
Page 44
Page 45
Page 46
Page 47
Page 48
Page 49
Page 50
Page 51
Page 52
Page 53
Page 54
Page 55
Page 56
Page 57
Page 58
Page 59
Page 60
Page 61
Page 62
Page 63
Page 66
Page 65
Page 66
Page 67
Page 68
Page 69
Page 70
Page 71
Page 72
Page 73
Page 74
Page 75
Page 76
Page 77
Page 78