Search engine with user activity memory

• US 6,490,577 B1
• Filed: 03/31/2000
• Issued: 12/03/2002
• Est. Priority Date: 04/01/1999
• Status: Expired due to Term

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Polyvista Incorporated
• Inventors: Anwar, Mohammed S.
• Primary Examiner(s): Nguyen, Cam Linh T.
• Application Number:
  US09/541,457
• Time in Patent Office: 977 Days
• Field of Search: 707/4, 707/3, 707/10, 705/10, 709/224
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/3338 Query expansion
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

A search engine is disclosed that utilizes both record based data and user activity data to develop, update and refine ranking protocols and to identify words and phrases that give rise to search ambiguity so that the engine can interact with the user to better respond to user queries and enhance data acquisition from databases, intranets and internets.

Claims

1. A search engine implemented on a distributed network comprising:

a. a user interface or a graphics user interface;

b. a packet processing module that converts each packet to at least one transaction and processes the transaction;

c. a database including records, keywords, user activity data (UAD), and a UAD updating routine for updating UAD;

d. a ranking protocol including record specific data (RSD) and UAD for ranking retrieved records from the records database;

e. communication and I/O,

where the ranking protocol comprises;

$\begin{array}{cc}r=\underset{i}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_i<br><br>w_i^{\mathrm{pi}}+\underset{i,i^{\mathrm{\prime <br><br>}}}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\beta <br><br>}}_{i,i^{\mathrm{\prime <br><br>}}}·<br><br>\left(w_i·<br><br>w_{i^{\mathrm{\prime <br><br>}}}\right)+\underset{j}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_j<br><br>w_j^{\mathrm{pj}}+\underset{j,j^{\mathrm{\prime <br><br>}}}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\beta <br><br>}}_{j,j^{\mathrm{\prime <br><br>}}}·<br><br>\left(w_j·<br><br>w_{j^{\mathrm{\prime <br><br>}}}\right)+\underset{i,j}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\gamma <br><br>}}_{i,j}·<br><br>\left(w_i·<br><br>w_j\right)& \left(1\right)\end{array}$

where r is a ranking factor, α

_i is a weight factor representing an amount of weight to be given to w_i which is a value of an i^th record specific criterion, p_i is a power factor of the i^th record specific criterion and i is an integer ranging over the record specific criterion and α

_j is a weight factor representing an amount of weight to be given to w_j which is a value a j^th user activity criterion, p_j is a power factor of the j^th user activity criterion and j is an integer ranging over the user activity criterion.

2. The search engine of claim 1, wherein the database module further includes:

records, keywords, and a UAD updating routine for updating UAD.

3. The search engine of claim 1, wherein the database module further includes:

at least one records database;

a keywords database including user activity data (UAD); and

a UAD updating routine for updating UAD.

4. The engine of claims 1, further comprising a query refinement module for identifying clustering of records based on UAD and RSD so that user queries can have improved relevance and context and artificial intelligence module for refining the ranking protocol, determining clustering, and UAD updating.

5. A search engine component comprising:

a. a packet processing module that converts each packet to at least one transaction and processes the transaction;

b. a database including user activity data (UAD); and

c. a ranking protocol including record specific data (RSD) and UAD for ranking retrieved records from the records database, where the ranking protocol comprises;

$\begin{array}{cc}r=\underset{i}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_i<br><br>w_i^{\mathrm{pi}}+\underset{i,i^{\mathrm{\prime <br><br>}}}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\beta <br><br>}}_{i,i^{\mathrm{\prime <br><br>}}}·<br><br>\left(w_i·<br><br>w_{i^{\mathrm{\prime <br><br>}}}\right)+\underset{j}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_j<br><br>w_j^{\mathrm{pj}}+\underset{j,j^{\mathrm{\prime <br><br>}}}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\beta <br><br>}}_{j,j^{\mathrm{\prime <br><br>}}}·<br><br>\left(w_j·<br><br>w_{j^{\mathrm{\prime <br><br>}}}\right)+\underset{i,j}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\gamma <br><br>}}_{i,j}·<br><br>\left(w_i·<br><br>w_j\right)& \left(1\right)\end{array}$

where r is a ranking factor, α

_i is a weight factor representing an amount of weight to be given to w_i which is a value of an i^th record specific criterion, p_i is a power factor of the i^th record specific criterion and i is an integer ranging over the record specific criterion and α

_j is a weight factor representing an amount of weight to be given to w_j which is a value a j^th user activity criterion, p_j is a power factor of the j^th user activity criterion and j is an integer ranging over the user activity criterion.

6. The component of claim 5, wherein the database module further includes:

records, keywords, and a UAD updating routine for updating UAD.

7. The component of claim 5, wherein the database module further includes:

at least one records database;

a keywords database including user activity data (UAD); and

a UAD updating routine for updating UAD.

8. The component of claim 7, wherein the transaction processing module for processing transactions includes:

a new query processing routine for retrieving records satisfying the query and ranking the retrieved records according to the ranking protocol;

a selected record processing routine including a first UAD activity value and a count incrementer,

a record range processing routine including a second UAD activity value and the count incrementer for identifying retrieved records viewed by a user;

a terminal activity processing routine including a third UAD activity value and the count incrementer.

9. The component of claim 5, further comprising a query refinement module for identifying clustering of records based on UAD and RSD so that user queries can have improved relevance and context and artificial intelligence module for refining the ranking protocol, determining clustering, and UAD updating.

10. A ranking protocol implemented in a digital processing unit comprising:

$\begin{array}{cc}r=\underset{i}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_i<br><br>w_i^{\mathrm{pi}}+\underset{i,i^{\mathrm{\prime <br><br>}}}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\beta <br><br>}}_{i,i^{\mathrm{\prime <br><br>}}}·<br><br>\left(w_i·<br><br>w_{i^{\mathrm{\prime <br><br>}}}\right)+\underset{j}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\alpha <br><br>}}_j<br><br>w_j^{\mathrm{pj}}+\underset{j,j^{\mathrm{\prime <br><br>}}}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\beta <br><br>}}_{j,j^{\mathrm{\prime <br><br>}}}·<br><br>\left(w_j·<br><br>w_{j^{\mathrm{\prime <br><br>}}}\right)+\underset{i,j}{\sum <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\gamma <br><br>}}_{i,j}·<br><br>\left(w_i·<br><br>w_j\right)& \left(1\right)\end{array}$

where r is a ranking factor, α

_i is a weight factor representing an amount of weight to be given to w_i which is a value of an i^th record specific criterion, p_i is a power factor of the i^th record specific criterion and i is an integer ranging over the record specific criterion and α

_j is a weight factor representing an amount of weight to be given to w_j which is a value a j^th user activity criterion, p_j is a power factor of the j^th user activity criterion and j is an integer ranging over the user activity criterion.