ended)
Q:
...
A: An Office365 plugin is available as well.
Q: How do you define “turn around time”? Does the agent need to communicate with the server every time it needs to classify a file?
A: That is correct, data travels to the Getvisibility Classification Server for the classification with a pre- configured frequency.
Q: What happens if an endpoint is disconnected or unable to reach the classification server?
A: In that case the user has to manually classify a file. After the agent connects back to the server the agent sends all the events that happened while offline.
Q: Can customers add their own PII definitions from the dashboard/ wizard (more like regex definitions to suggest a particular classification)?
A: Yes that will be possible for the customers to configure their regexes.
Q: Is there integration into DLP Products so we can do automated labeling from an Endpoint Discovery Task?
A: There will be integration in the future.
Q: Why is GV Classification superior to existing vendor?
A: There are a number of areas where this new classification solution is even better. The main advantage is that GV Classification effectively leverages artificial intelligence (AI) and machine learning (ML) to provide superior classification that is industry and business specific.
Q: Is the ML system trained with customer data, public data or is it just initial base data?
A: The ML model is based on real business data from organizations. Deployments start with a master model and then it adapts to the customer’s environment. Customers who participate to the Feature Store increase the intelligence of the ML and so it also continuously improves accuracy.
Q: Is there large sample sets available for demo/POC where a customer does not want to use production data?
A: Yes – and it will also improve as the organizations use the product.
Q: Any plans for adding support for AutoCAD?
A: Yes – an AutoCAD plug-in is in the roadmap.
Q: Does the user have to manually add the tags each time? How much automation can be configured? A: We provide AI suggestions to help end-users to apply the classification tag. Default auto- labelling is a also an option that user can leverage.
Q: How will this suggested popup behave? Only on opening a file? Or will it popup several times when a user is creating a file and adding information to it either through copy/paste or just typing?
A: We have config for how often to show a pop-up or not to show it at all. In production we do not activate the suggestion pop-up. When a user is ready to classify a file, the suggestion is already there for them. That is a configurable option.
Q: Do you save any of my data?
A: No, we never store any of your file content. The classification server maintains a registry of file names and their properties but not the content. We have even built an anonymization mechanism into the GV Classification software that reduces file content to a mathematical number that is used throughout the platform.
Q: Will this affect my network or file server performance?
A: No, the software runs in a throttled manner that controls the rate at which we scan files. We appear like a normal staff workstation. On the staff laptops or desktops we run very lightweight plugins to interact with the staff member, suggest document classifications and alert the staff member to risky actions.
Q: Does your software only handle files and emails in English?
A: No, there is support for multiple languages with English as the standard deployment/default language. We have additional language options available for German, French, Spanish, Italian and also Arabic. Chinese and Thai variants are planned for the near future.
Q: Do we have to give you sample data?
A: Usually no data is needed from the customer. If our AI model reports new document types it has not seen before, we request a small sub-set of sample data (a few hundred files) of these new files, which are immediately converted to an anonymous descriptive number we then use to train and update our model. This will ultimately help in improving the accuracy of our classification results. This process uses none of the actual document data.
Q: What is the ML model?
A: We use a combination of AI techniques that analyze document content and suggest descriptive tags to users during manual classification or during an automated scan. These AI techniques are based on Natural Language Processing (NLP) and neural networks. The software has been trained for more than 3 years and the package of information that is distributed to our servers containing this knowledge is called the model.
Q: How can I manually modify a document or email with GV Classification?
A: The agent software gives a list of the available compliance, classification, and possible categorization tags active within your organization. When you choose the relevant ones for the document or email in creation, the software will modify and track that document or email in the future. If staff remove the automatic visual tags, the document continues to keep the metadata tags and the centralized audit log keeps a record of the classification.
Q: How does the ML model help with manual classification?
A: A staff member classifies a document or email with the classification agent by selecting tags for compliance and classification. We also provide a suggested set of tags based on the actual content of the document or email. Staff can choose to use suggested tags or set completely different ones. If they select different tags the GV Classification software evaluates its own knowledge of the document and decides to either learn from the new tags or to generate an audit log if we believe the staff member has made a mistake. This allows for training and identifies staff errors, but crucially enables the GV Classification software to learn from expert users.
Q: How does GV Classification teach staff about better data hygiene?
...
What is the ML model?
A: We use a combination of AI techniques that analyze document content and suggest descriptive tags to users during manual classification or during an automated scan. These AI techniques are based on Natural Language Processing (NLP) and neural networks. The software has been trained for more than 3 years and the package of information that is distributed to our servers containing this knowledge is called the model.
Q: How can I manually modify a document or email with GV Classification?
A: The agent software gives a list of the available compliance, classification, and possible categorization tags active within your organization. When you choose the relevant ones for the document or email in creation, the software will modify and track that document or email in the future. If staff remove the automatic visual tags, the document continues to keep the metadata tags and the centralized audit log keeps a record of the classification.
Q: How does the ML model help with manual classification?
A: A staff member classifies a document or email with the classification agent by selecting tags for compliance and classification. We also provide a suggested set of tags based on the actual content of the document or email. Staff can choose to use suggested tags or set completely different ones. If they select different tags the GV Classification software evaluates its own knowledge of the document and decides to either learn from the new tags or to generate an audit log if we believe the staff member has made a mistake. This allows for training and identifies staff errors, but crucially enables the GV Classification software to learn from expert users.
Q: How does GV Classification teach staff about better data hygiene?
A: GV Classification interacts with staff while they work on documents and emails. When classifying these, we present suggestions and also block certain risky actions such as sending internal documents outside the company. Whenever a staff member is blocked or warned we can present a text summary on request that explains why this action was blocked or why the classification is needed. We also explain where to find more information and even indicate the relevant internal policy and procedure. Effectively, we reinforce their data security training as they work.
Q: The demo during training is based on the office apps installed on an endpoint. Is there also support for the online versions of the office suite? For example, Office365 word in a browser?
A: An Office365 plugin is available as well.
Q: How do you define “turn around time”? Does the agent need to communicate with the server every time it needs to classify a file?
A: That is correct, data travels to the Getvisibility Classification Server for the classification with a pre- configured frequency.
Q: What happens if an endpoint is disconnected or unable to reach the classification server?
A: In that case the user has to manually classify a file. After the agent connects back to the server the agent sends all the events that happened while offline.
Q: Can customers add their own PII definitions from the dashboard/ wizard (more like regex definitions to suggest a particular classification)?
A: Yes that will be possible for the customers to configure their regexes.
Q: Is there integration into DLP Products so we can do automated labeling from an Endpoint Discovery Task?
A: There will be integration in the future.
Q: Why is GV Classification superior to existing vendor?
A: There are a number of areas where this new classification solution is even better. The main advantage is that GV Classification effectively leverages artificial intelligence (AI) and machine learning (ML) to provide superior classification that is industry and business specific.
Q: Is the ML system trained with customer data, public data or is it just initial base data?
A: The ML model is based on real business data from organizations. Deployments start with a master model and then it adapts to the customer’s environment. Customers who participate to the Feature Store increase the intelligence of the ML and so it also continuously improves accuracy.
Q: Is there large sample sets available for demo/POC where a customer does not want to use production data?
A: Yes – and it will also improve as the organizations use the product.
Q: Any plans for adding support for AutoCAD?
A: Yes – an AutoCAD plug-in is in the roadmap.
Q: Does the user have to manually add the tags each time? How much automation can be configured?
A: We provide AI suggestions to help end-users to apply the classification tag. Default auto- labelling is a also an option that user can leverage.
Q: How will this suggested popup behave? Only on opening a file? Or will it popup several times when a user is creating a file and adding information to it either through copy/paste or just typing?
A: We have config for how often to show a pop-up or not to show it at all. In production we do not activate the suggestion pop-up. When a user is ready to classify a file, the suggestion is already there for them. That is a configurable option.
Q: Do you save any of my data?
A: No, we never store any of your file content. The classification server maintains a registry of file names and their properties but not the content. We have even built an anonymization mechanism into the GV Classification software that reduces file content to a mathematical number that is used throughout the platform.
Q: Will this affect my network or file server performance?
A: No, the software runs in a throttled manner that controls the rate at which we scan files. We appear like a normal staff workstation. On the staff laptops or desktops we run very lightweight plugins to interact with the staff member, suggest document classifications and alert the staff member to risky actions.
Q: Does your software only handle files and emails in English?
A: No, there is support for multiple languages with English as the standard deployment/default language. We have additional language options available for German, French, Spanish, Italian and also Arabic. Chinese and Thai variants are planned for the near future.
Q: Do we have to give you sample data?
A: Usually no data is needed from the customer. If our AI model reports new document types it has not seen before, we request a small sub-set of sample data (a few hundred files) of these new files, which are immediately converted to an anonymous descriptive number we then use to train and update our model. This will ultimately help in improving the accuracy of our classification results. This process uses none of the actual document data.
Q: Can GV Classification use special keywords to classify documents and email?
...