The present disclosure relates generally to a manufactured part archive and, more particularly, to a method for linking information to the archive.
Knowledge archives may store textual and/or graphical information. Textual information may often be effectively searched using keywords. When using textual keyword searching, users may often input a series of keywords and Boolean operators. A variety of search algorithms exist that are designed to discover the information most related to the search keywords, but even the most efficient algorithms may be mottled with shortcomings. Textual database searching is often overbroad whereby the user is presented with irrelevant results; or is often overly exclusive whereby the algorithm fails to find all relevant information. These shortcomings may be exacerbated when using textual keywords to search non-textual or graphical data.
As such, it would be desirable to provide an improved method for linking, searching, and retrieving information stored in a graphical archive.
A method of linking information to an electronically enabled manufactured part archive may include identifying an area of the electronically enabled manufactured part archive, defining the area with a geometric token, and electronically identifying information. The information may relate to at least a portion of a manufactured part, which is associated with the identified area of the manufactured part archive. The method may also include electronically associating the information with at least one of the identified area or the geometric token, wherein the information is adapted for retrieval.
Objects, features and advantages of embodiments of the present disclosure may become apparent by reference to the following detailed description and drawings, in which like reference numerals correspond to similar, though not necessarily identical, components. For the sake of brevity, reference numerals having a previously described function may not necessarily be described in connection with other drawings in which they appear.
A method of linking and retrieving information from an archive is provided whereby information and/or data may be linked with a finite location in an archive record. As such, a user may be able to search for textual information associated with such a finite area. Furthermore, a user may be able to view all information linked with a defined area of an image or to search within the information linked with a particular finite area. Such a system may advantageously provide an intuitive, visual-based means for improved linking and retrieval of information within a knowledge archive.
Referring to
Referring now to
It is to be understood that, in any of the embodiment(s) discussed herein, the geometric token 32 may be representative of a class of information. As such, each of a plurality of geometric tokens 32 may represent a different class of information and may be distinguished via geometric token 32 shape, geometric token 32 size, and/or geometric token 32 color. It is to be understood that, as defined herein, “color” is intended to include any of fill color, fill patterns, fill shading, border color, border patterns, border shading, or the like, or combinations thereof. A non-limitative example of patterns includes cross hatching.
As a non-limitative example embodiment, a round geometric token 32 may identify a manufacturing problem relating to paint, a square geometric token 32 may identify a manufacturing problem relating to welding, and/or a hexagonal geometric token 32 may identify a manufacturing problem relating to a stamping issue. Thus, visually distinct geometric tokens 32 may be used to identify the class of information linked to an area 24. It may be preferred to identify a class of information when linking information to an area 24 to provide a way to later retrieve information related to only a specified class of information when searching within an EEMPA 12.
In another embodiment, a plurality of areas 24 may be identified, as depicted at reference numeral 28, and each of the areas 24 defined, as depicted at reference numeral 36, with a respective geometric token 32. Further, the information may be substantially simultaneously or sequentially electronically associated with each of the plurality of identified areas 24, each of the plurality of geometric tokens 32, and/or combinations thereof. Thus, a single piece of information may be effectively and efficiently associated with a plurality of areas 24 and/or geometric tokens 32.
Referring now to
In an embodiment of method 48 including a plurality of geometric tokens 32, each token 32 may represent a different class of information, and each of the plurality of geometric tokens 32 may be distinguished via at least one of geometric token 32 shape, geometric token 32 size, and/or geometric token 32 color.
Referring again to
In an embodiment, the identified area 24 has a common feature with another area 24. A common feature may, for example, be a component or subcomponent of a manufactured part, which is referenced in two or more records 16. In another embodiment, a common feature may be a feature having a common purpose and/or design to another non-identical feature. For example, in an EEMPA 12 containing records relating to vehicles, a common feature may exist between two records 16 representative of components designed for the same function in different vehicle models. In an embodiment wherein an identified area 24 has a common feature with another area 24, retrieving information, as depicted at reference numeral 60, may further include retrieving information electronically associated with the other area 24.
Each record 16 may be a subcomponent of, and/or include a common feature with a vertically adjacent record 16. Thus, referring to
In an embodiment of linking information, each associated area 64, 68, 72, 76 is effectively defined, as depicted at reference numeral 36, and is electronically associated with the information, as depicted at reference numeral 44. In an embodiment of retrieving information, information electronically associated with each associated area 64, 68, 72, 76 may be retrieved, even if the information was previously electronically associated with less than all of the associated areas 64, 68, 72, 76.
A plurality of records 16 may be organized in a hierarchical structure having at least one of vertically or horizontally adjacent records 16, as illustrated in the EEMPA 12 embodiment of
In an embodiment of a hierarchical structure, substantially each record 16 is a subcomponent of a vertically or horizontally adjacent record 16. It is to be further understood that “substantially” each record 16 may contain a subcomponent because at least one record 16 in such a hierarchical arrangement may contain an assembly of subcomponents rather than a subcomponent. For example, record 18 of
Referring still to
It is to be understood that the EEMPA 12 shown in
As a non-limitative example, a user may desire to link data related to manufacturing issues occurring at a “top front portion of a door frame.” The user may electronically define, as depicted at 36, the area 24 (in this example, the “top front portion of a door frame”) with a geometric token 32, such as a circle (as illustrated at reference numeral 68) in record 18′. The user may then electronically identify the information related to the manufacturing issues, and electronically associate the information with the identified area 24.
Referring additionally now to
In an embodiment, the geometric token 32 is defined and placed via a graphical user interface (GUI). The parameters defining the geometric token 32 may be expressed in the record's own coordinate system. A non-limitative example of such parameters adapted to define a circle include: X-coordinate, Y-coordinate, and radius.
In
Assessing a similarity between two or more areas 24, or similarity matching, may be used to compute a similarity between two or more areas 24. Furthermore, similarity matching may provide a means to determine whether a first defined area 24 includes a common feature with one or more other defined areas 24. As such, similarity matching may be helpful in both linking information to, and retrieving information from an EEMPA 12. For example, in an embodiment of linking information to an EEMPA 12, similarity matching may be performed before electronically associating the information with the defined area 24, such that the information is effectively electronically associated with the area 24 of each record 16 having a common feature with the defined area 24. In another example of an embodiment, similarity matching may be used after the step of identifying the area 24, such that information is retrieved from each area 24 of a record 16 having a common feature with the defined area 24.
Similarity comparisons between each of the defined areas 128, 132, 136 of the records 116, 120, 124 and the defined area 104 of the example record 100 are shown as
The similarity between the example record 100 and Case 1 (reference numeral 116) is relatively high because the area of overlap 152 between the areas 104, 128 is high with respect to the area of non-overlap 156. The similarity between the example record 100 and Case 3 (reference numeral 124) is lower than that of the example record 100 and Case 1 (reference numeral 116) because the area of overlap 152 for Case 3 (reference numeral 124) is smaller with respect to the area of non-overlap 156 for the areas 104, 136. Finally, the similarity between the example record 100 and Case 2 (reference numeral 120) is even lower than that of Case 3 (reference numeral 124) because there is no overlap between the areas 104, 132.
A non-limitative embodiment of a method of assessing a similarity between a first defined area 24 and at least one other defined area 24 in an EEMPA 12 having at least one record 16 includes assessing an amount of overlap between the first defined area 24 and the other defined area(s) 24, the first defined area 24 having a common feature with the other defined area(s) 24. The method further includes normalizing the amount of overlap between the first defined area 24 and the other defined area(s) 24. The amount of overlap may be normalized by the surface areas of the first defined area 24 and the other defined area(s) 24. In an embodiment, the similarity is a number ranging from 0.0 to 1.0.
The method of assessing a similarity may further optionally include the step of organizing the plurality of records 16 such that each of the other defined area(s) 24 is electronically associated with the first defined area 24.
In an embodiment, when an area 24 is electronically associated with information, as depicted at 44, the area 24 is “inherited” through the “part of relation by levels 80, 84, 86 above and below in the appropriate taxonomy. As such, information linked to an area 24 is effectively linked to other area(s) 24 sharing a common feature. Thus, each linked record 16, Q, is associated with a collection of linked records 16, C(Q), that represent the same location. C(Q) may be referred to as an “associated record set” for Q. It is to be noted that Q is a linked record 16 and not just a record 16. As such, C(Q) depends on the linking, as well as the record 16 that was originally linked.
With a slight abuse of notation, Q can be thought to indicate the linked area 24 of the record 16. For another linked record 16, T, to have a non-zero similarity to Q, one of the records 16 in C(Q) must be the same as (or include a common feature with) one of the records 16 in C(T). The similarity of Q to T, s(Q,T), is a number from 0.0 to 1.0. Thus, s(Q,T)=0.0 means that the locations indicated by Q and T are not close to each other; while s(Q,T)=1.0 indicates that Q and T are essentially the same. As such, s(Q,T) will generally be a number less than 1.0.
It is to be understood that the similarity function “s” need not be symmetric in its arguments. As such, if Q is the “query” (or area 24 defined for information retrieval), and T is the “target” (or area 24 linked to desired information), it is reasonable for s(Q,T) to be close to 1.0 when Q contains T, but a lesser value when T contains Q.
In an embodiment, s(Q,T) is computed from the amount of overlap of the areas 24 indicated by Q and T, normalized by the surface areas of Q and T. If either Q or T is a point, and, thus, has no area, then s(Q,T) may be max[0.0,(Dthres−the distance from T to Q)/Dthres]. Dthres is defined as the maximum distance between T and Q that will result in a similarity>0.0). Otherwise, a formula such as s(Q,T)=area(Q intersect T)/(square root [area (Q)*area (T)]) could be used to find the similarity of Q and T. Using this formula, s(Q,T)=1.0 if Q=T and s(Q,T)=0.0 if (Q intersect T) is empty. Otherwise, the similarity is a number between 0.0 and 1.0.
While several embodiments have been described in detail, it will be apparent to those skilled in the art that the disclosed embodiments may be modified. Therefore, the foregoing description is to be considered exemplary rather than limiting.
This application is a divisional of co-pending U.S. Ser. No. 11/536,002, filed on Sep. 28, 2006, which is incorporated by reference herein in its entirety.
Number | Date | Country | |
---|---|---|---|
Parent | 11536002 | Sep 2006 | US |
Child | 12697928 | US |