Method and device for automatically issuing information using a search engine

• US 7,167,545 B2
• Filed: 11/27/2001
• Issued: 01/23/2007
• Est. Priority Date: 12/06/2000
• Status: Expired due to Term

• Current Assignee: Volt Delta International GmbH (Volt Information Sciences Incorporated)
• Original Assignee: varetis solutions GmbH (Volt Information Sciences Incorporated)
• Inventors: Dahmen, Michael, Wagner, Johannes, Heidenfelder, Klaus, Plannerer, Bernd
• Primary Examiner(s): Gauthier; Gerald
• Application Number:
  US10/433,888
• Publication Number:
  US 20040022372A1
• Time in Patent Office: 1,883 Days
• Field of Search: 379/88.01, 379/88.02, 379/88.04, 704/231, 704/246, 704/261
• US Class Current: 379/88.01
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

Information sought, with associated attributes, is stored in the form of data records. Embodiments provide for inquiry of search arguments for several attributes stored in a data record; comparison of the input search arguments with search arguments stored; selection of a number of hits of storage search arguments corresponding to the respective input search argument, for each of the search argument inputs; weighting of the selected search arguments with scores, which weighting indicates the probability with which the respective selected stored search argument agrees with the actually input search argument; selection of suitable data records from the database via the selected number of hits; weighting of the selected data records with overall scores indicating the probability of the respective selected data record agreeing with the actually input search arguments, depending on the scores of the individual selected search arguments; and output of the data record with the highest overall score.

Claims

1. A method for automatically providing information via a search engine, the method comprising:

receiving search arguments for a plurality of attributes stored in a data record, the plurality attributes comprising a name and a location of a searched object;

comparing of the received search arguments with stored search arguments;

for each of the received search arguments, selecting at least one of stored search argument hit that approximates the received search argument;

weighting the at least one selected search argument hit;

wherein each weight indicates a probability that a selected stored search argument agrees with a respective received search argument;

selecting data records from the stored data records using the selected at least one search argument hit;

weighting the selected data records with overall scores indicating a probability of a respective selected data record agreeing with the received search arguments; and

wherein the overall scores depend on the scores of the at least one selected search argument hit.

2. The method according to claim 1, wherein the step of receiving the search arguments comprises speech recognition.

3. The method according to claim 1, wherein the step of receiving the search arguments takes place step-by-step for each attribute separately.

4. The method according to claim 1, wherein:

a plurality of separate quantities of stored search arguments are provided according to various types of attributes; and

the step of comparing the received and stored search arguments for a respectively received search argument in each instance is carried out on the basis of only one of the quantities of stored search arguments.

5. The method according to claim 1, wherein the step of selecting data records comprises selecting data records that contain attributes that correspond to one of the search arguments of each of the at least one search argument hit.

6. The method according to claim 1, wherein the step of selecting data records comprises selecting data records that comprise the largest number of attributes that correspond to a respective search argument of each of the at least one search argument hit.

7. The method according to claim 1, wherein:

the overall scores are determined by multiplying individual scores allocated to the attributes of respective search arguments; and

the individual scores for attributes of the respective search arguments are weighted individually.

8. The method according to claim 1, wherein:

for selected search arguments for at least one of the attributes a check is generated which reproduces the selected search argument for a respective attribute with the best score; and

a number of hits of the selected search arguments for the respective at least one attribute during confirmation of the check is limited to the search argument with the best score.

9. The method according to claim 1, further comprising:

outputting the data record with the highest overall score; and

announcing a telephone number corresponding to the output data record.

10. The method according to claim 1, wherein the name is weighted, the location is weighted, and the weights of the name and location are multiplied to calculate the overall score.

11. The method of claim 1, further comprising outputting the data record with the highest overall score.

12. The method according to claim 11, wherein output of the data record takes place acoustically via a speech reproduction unit.

13. The method according to claim 11, wherein a plurality of data records are output according to their overall scores.

14. The method according to claim 13, wherein:

the step of selecting a data record comprises querying a further attribute; and

an entered search argument of the Queried attribute is only compared to attribute values of the data records already selected.

15. The method according to claim 11, wherein the steps are carried out with a single search engine to automatically output a single data record with highest overall score.

16. A method for automatically providing information by means of a search engine, in which information sought, together with other associated attributes, is stored in the form of data records, involving the following steps:

inquiry of search arguments for several attributes stored in the data record;

comparison of the search arguments which have been input with search arguments stored;

selection of a number of hits of stored search arguments which correspond to, or most closely approximate, the respective search argument that has been input, for each of the search arguments input;

weighting of the selected search arguments with scores, which weighting indicates the probability with which the respective selected stored search argument agrees with the actually input search argument;

selection of suitable data records from the stored data records by means of the selected number of hits;

weighting of the selected data records with overall scores which indicate the probability of the respective selected data record agreeing with the actually input search arguments, depending on the scores of the individual selected search arguments; and

output of the data record with the highest overall, wherein the search arguments which have been selected are weighted with absolute scores, wherein, preferably, a score is specified and if the scores are outside this score range, steps a) to d) are repeated in relation to the respective search argument.

17. A device for automatically providing information via a search engine the method comprising

an input device for inputting search arguments for a plurality of attributes stored in a data record;

wherein the plurality of attributes include a name and a location of a searched object;

a comparison device for comparing the input search arguments with stored search arguments;

a selection device for selecting at least one stored search argument hit that approximates an input search argument for each of the input search arguments that have been input;

a weighting device for weighting the selected at least one search argument hit;

wherein each weight indicates a probability a selected stored search argument agrees with a respective input search argument;

a selection device for selecting data records from the stored data records using the at least one search argument hit;

a calculation device for calculating overall scores for the selected data records indicating a probability of the a respective selected data record agreeing with the input search arguments; and

wherein the overall scores depend on the scores of the at least one selected search argument hit.

18. The device according to claim 17, further comprising a speech recognition device for speech recognition of spoken input search arguments.

19. The device according to claim 17, further comprising an inquiry device configured such that the input search arguments are queried step-by-step for each attribute separately.

20. The device according to claim 17, further comprising:

a memory connectable to the speech recognition device

wherein several separate quantities of stored search arguments are stored in the memory according to various types of attributes;

wherein the comparison device is configured such that a comparison between an input and a stored search argument for a respectively input search argument is carried out on the basis of only one of the quantities of stored search arguments.

21. The device according to claim 17, wherein the weighting device is designed such that absolute scores are allocated to the selected at least one search argument.

22. The device according to claim 17, wherein the selection device is configured to select data records that contain attributes that correspond to one of the search arguments of each of the at least one search argument hit.

23. The device according to claim 17, wherein the selection device is configured to select data records that comprise the largest number of attributes that correspond to a respective search argument of at least one search argument hit.

24. The device according to claim 17, wherein:

the calculation device is configured such that the overall scores are determined by multiplying individual scores allocated to attributes of respective search arguments; and

the calculation device configured such that scores for attributes are weighted individually.

25. The device according to claim 17, further comprising:

a checking device for generating a check, which in relation to a group of selected search arguments, returns the search argument with the best score;

wherein the selection device is configured such that during confirmation of the check the at least one search argument hit is limited to the search argument with the best score.

26. The device according to claim 17, further comprising an output device comprising a speech reproduction unit for acoustic output.

27. The device according to claim 17, wherein the device is adapted to operate as a telephone-number information system.

28. The device according to claim 17, wherein the device is adapted to access a single search engine and output a single data record with a highest overall score.

29. The device according to claim 17, wherein the weighting device weights the name, separately weights the location, and the calculation device multiplies the weights of the name and location to calculate the overall score.

30. A computer program product comprising program code, stored on a computer-readable data carrier, for implementing the following steps;

receiving search arguments for a plurality of attributes stored in a data record, the plurality of attributes comprising a name and a location of a searched object;

comparing the received search arguments with stored search arguments;

for each of the received search arguments, selecting at least one stored search argument;

hit that approximates the received search argument;

weighting the at least one selected search argument hit;

wherein each weight indicates a probability that a selected stored search argument agrees with a respective received search argument;

selecting data records from the stored data records using the selected at least one search argument hit;

weighting the selected data records with overall scores indicating a probability of a respective selected data record agreeing with the received search arguments; and

wherein the overall scores depend on the scores of the at least one selected search.