Browse Prior Art Database

Automated search of social file share for document referenced by natural language

IP.com Disclosure Number: IPCOM000242946D
Publication Date: 2015-Sep-01
Document File: 2 page(s) / 28K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system that automates the search of the social file share in order to find links to documents being referenced in the user's correspondence, such as an email message. The system takes the natural language description of a document (for example, "last month's Equity Strategies document"), as well as messaging metadata, and automates a search for the document without requiring input from the user, offering links to the most probable matches for this document in the social file storage.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 49% of the total text.

Page 01 of 2

Automated search of social file share for document referenced by natural language

Disclosed is a system that searches the social file share storage to identify links to documents being referenced in the user's correspondence; for example, documents mentioned in the user's received email. The system can take the natural language description of a document such as "last month's Equity Strategies document", as well as messaging metadata, and offer links to the most probable matches for this document in the social file storage, without requiring input from the user.

    Business messages and emails frequently discuss topics which reference supporting information, such as spreadsheets, presentations, and other documents. For example, the sender's email says, "Please provide your comments on the issues listed in last month's Equity Strategies document." Immediately the recipient (the user) is asking himself where to find this document.

    The message sender and the receiving user may belong to communities where shared files are posted in the social file store. In this case, the sender might fail to send a link to the referenced document, assuming the user knows where to find the referenced information in a social file store. However, the location of the file in the social file store may not be obvious to the user. For example, would the Equity Strategies document likely be found in the financial community, or in the sales community, or in the marketing community? It can be difficult for the user to keep track of where to find a particular file; therefore, the user has a great advantage when there is a system to automatically locate the file for opening it or downloading it.

    The disclosed system can automate a search for the file in the social space, by creating search inputs based on the natural language in the message, as well as utilizing messaging metadata. Since the disclosed system automates the search, the message sender need not explicitly provide a link to a referenced document. The user is relieved of the burden to consult bookmarks to the social file share communities, to see what has been deposited in the any of the communities. The user need not conduct a global search across the social communities to locate the supporting information using keywords, contact names, or file names. The disclosed system does not require any extra effort by the user, or by the sender of the message.

    In the disclosed system, natural language in the message can be leveraged to provide
clues of what to search for. Without any user inputs (other than perhaps a user-defined preference to search for referenced documents), the system locates potential matches in the social file share for the document referenced in the current message. Natural language processing (NLP) systems, such as IBM Watson*, are enabling technology for the disclosed system, used to understand the text in the message. The NLP deconstructs a sentence to locate any reference to a document,...