Browse Prior Art Database

Method and System of Personal Information Demands Tracking on the Web Disclosure Number: IPCOM000198087D
Publication Date: 2010-Jul-26
Document File: 1 page(s) / 78K

Publishing Venue

The Prior Art Database


In the past years, the rapid development of online information has aroused a lot of studies from the industries and businesses, mainly due to its tremendous value for potential knowledge exploration. Communications between people within online communities is one of the most important information exploited from web 2.0 and represent a new path towards providing continuous information about user personal information demands status. Traditional approaches for user demand exploration on web are usually static without considering time factor. In fact, user’s demands on information change dynamically, especially along with their discussions with other people. The influence existing between users is obvious to change user’s idea and knowledge. It is necessary to detect such kind of change when exploring user’s demands on specific product or something else. A lot of potential applications can be implemented based on such kind of information discovery, such as, setup user personal profile for web advertising and product recommendation etc. Problem Solved: Within web forum style discussion, users seek others’ support, advice and information about some specific kind of object. The messages of users can be considered as the unstructured diaries detailing user’s requirement, including the brand, the required function, the condition as well as user preference etc. Often the responses detail the suggestion from others according to their individual experience and knowledge. These responses will have more or less effect on others’ idea, which can be reflected by user’s following message. Below is an example to illustrate this scenario. User A: I want to purchase a camera, who guys can give me some suggestions? User B: amateur or professional? User A: amateur, I like travel, and want to have a camera easy for use. It is not necessary to have high pixel, 500MEGA is ok. User B: a lot of people like travel select professional camera. User A: no, I want a camera easy for use. User B: oh, I see. I think you should select a camera with compact size. User A: yes, you are right. It should be easy to carry. User C: I think you’d better select a camera with high pixel. 1000MEGA is the mainstream configuration. User A: really, but I worry the price is high. My budget is 2000-2500. User C: don’t worry about the price, price of a lot of camera with good configuration can be accepted. I suggest you to take a look at the camera A1, A2,… User A: maybe you are right, the high pixel is necessary to take picture when travelling… What demand information can be explored from the above conservation? A camera Amateur camera With compact size, lightweight with high pixel with reasonable price What’s demand is changed during the conservation? The pixel is changed from 500MEGA to 1000MEGA What’s demand is not changed during the conservation? Demand for amateur Camera What can be inferred from other clues? The user like travel, therefore the camera is better rainproof and shockproof. Most of traditional approaches ignore the change of demands caused by the influence between people and also have no the ability to infer the implicit demand specification from user’s description. In this invention, we investigate an efficient approach for demand tracking based on the conservations between users on web.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 63% of the total text.

Page 1 of 1

Method and System of Personal Information Demands Tracking on the Web

3.1 Overall Architecture

The overall architecture is displayed in the Figure 1.

 User explicit requirements on product features

 User explicit requirements on product features

 User implicit requirements on product features

3.2 The personal information demands tracking on the web method mainly consists of the following steps:

Step 1: Detect information demander and the demanded objects.

   Step 1.1: based on QA detection techniques, posts with implicit or explicit information requirements are extracted.

Step 1.2:

NER and NLP techniques are implemented to identify who issue the request and for

Step 2: Reply-to relationship identification. By analyzing the structure of the focused web page, identify the responses relationship between users.

Step 3: User requirement detection.

Step 3.1: explicit requirement detection

Step 3.1.1: identify object features explicated discussed by user. It is based on the exploration of features from product review or specifications.

      Step 3.1.2: identify the specific requirement on each identified feature. Here, time factor are considered in case the change of user's requirement on the same feature. The final status of user requirement on each object feature is determined after the analysis throughout all the discussion thread.

Step 3.2: implicit requirement detection

      Step 3.2.1: collect user related information as much as possible, such as, related event of the targeted ob...