Browse Prior Art Database

A method for automatic directory recommendation for storing downloads using structured output ranking Disclosure Number: IPCOM000232541D
Publication Date: 2013-Nov-15
Document File: 3 page(s) / 37K

Publishing Venue

The Prior Art Database


An automatic download directory recommender which is trained using the context information: The information associated with the usual habit of storing downloads, history information; Based on work focus and schedule, recommend directory based on the download file whether is related to current project; Analysis current html page and obtain the downloads’ related information such as file name, abstract information, introduction info. Then use semantic analysis to tag this file and recommend it to target directory; The same method can also be applied for assisting locating stored file, especially in cloud storage, searching a file from scratch is limited by network speed

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 57% of the total text.

Page 01 of 3

A method for automatic directory recommendation for storing downloads using structured output ranking

Different downloaded or transferred data should be stored in different directory based on different context:
- For different application,
- For downloading data using browser,
- For different type of data, such as doc, pdf, jpg.
- For documents with different content or topic.
- when work in different projects or different period.

When people store different downloaded data.
- people always jump target directory step by step
- store them in default directory, then copy to target directory
- Sometimes may not consider which directory to store, and just choose random directory, but after several days, do not know where the data is.
- Active directory in different period can be used frequently.

A method and system for recommending downloaded directory including two stages:
- Offline

• Input: the historical downloaded data, its context information, current page information and the associated stored directory

• Process: train the directory recommendation system for downloaded data.

- Online


Page 02 of 3

• Input: The current data information and context

• Output:

Recommended directory rank list based on its priority

Based on available historical data, train a recommender for every current different available directory:

1. Extract the set of following features:

- file title

- some info for downloaded file in current page if have

- file type such as exe, pdf, jpg

- file...