Employees often use a variety of applications and platforms in their daily work such as Microsoft Teams, Slack, calendars, and document management systems like SharePoint or OneDrive. These systems contain valuable insights about each employee, including their knowledge and expertise. Starmind uses advanced AI algorithms to access the information contained in these different systems to learn about each employee and build more accurate expertise profiles for each individual. More accurate expertise profiles mean team members can more quickly identify subject matter experts resulting in all-around better employee experience and higher productivity for the entire company.
Connectors should be enabled at the beginning of the Starmind lifecycle. Connectors eliminate the problem of a “cold start,” which happens when there is no information in the Starmind platform about a new user to build them an expertise profile. When connectors are enabled, Starmind’s algorithm can automatically create expertise profiles for every user whether or not they have already interacted on the Starmind platform.
Once a connector is set up, Starmind will run an initial learning phase and will then continue to learn from the information in this system on an ongoing basis, for example, when new posts are added to a workspace, or new messages are written in a communication channel.
Additional Starmind Connectors can be enabled at any point in time. Each additional connector will allow users to benefit from more accurate expert profiles of their colleagues.
With our calendar connectors, Starmind can learn from calendar events. Starmind will learn from all events with at least two participants and that are visible to all users. Starmind will not learn from private calendar entries and from calendar entries with one participant. Starmind processes only the title of the calendar entry, the description, and the list of participants. Starmind ignores all participants who do not have a Starmind account.
Chat & Collaboration Systems
Chat connectors allow Starmind to learn from chat messages that employees post in public channels in Microsoft Teams and Slack. Starmind does not learn from messages that were posted in protected channels and in private 1:1 channels. Starmind also ignores references and messages of people who do not have a Starmind account. Starmind processes only the text of the messages. Links and attachments are ignored.
Starmind determines the expertise of the users based on chat messages and assigns the corresponding expertise tags to the Starmind user profiles of the author of the message and of the people referenced in the message.
Starmind can learn the expertise of users from Salesforce Chatter. Starmind processes the text of the questions, answers, and posts provided in Salesforce Chatter by users and assigns the expertise tags to Starmind users corresponding to Salesforce Users referenced in the Salesforce Chatter.
With our storage connectors, Starmind can learn from the knowledge within files your employees create on a daily basis. We will compare the different versions of the files and link the knowledge we extract to the employee which did the change.
Social Collaboration Connectors
In social collaboration applications like Facebook Workplace, Starmind learns from the content posted by users. Starmind processes only the text of the posts published by users in the public groups (not protected and private groups). Starmind then assigns the extracted expertise tags to the Starmind user profiles corresponding to the user profile in the social collaboration application.
HR Information Systems
Human resources connectors provide essential information about employee's skills and knowledge. Starmind won't be able to read any private data from the human resources, only CVs
In ticketing tools like Jira, Starmind learns from the tickets created by users. Starmind processes from the title, short description, description, notes/comments, and closure/resolution information of tickets.
If the application is running on-premise, Starmind can't connect directly to the application. In that case, Starmind offers another solution to learning from such a system. More details around this approach can be found here: