Method and system for address result arbitration

Abstract

A method to arbitrate between results obtained from observationally derived data by several procedures, where the results are items characteristic of the observationally derived data is disclosed. Each procedure is given a ranking according to the confidence on the procedure. The results and characteristics derived from the results are used in a plurality of rules, where the rules are used to arbitrate between the results.

Description

BACKGROUND OF THE INVENTION

[0002] This invention relates generally to the arbitration between results obtained by more than one method for matching observationally derived data to data from a number of databases, and, more particularly to arbitrating or deciding between address recognition methods for automatically recognizing addresses on items in an automated delivery system.

[0003] Due to the ever increasing volume of postal items and packages being delivered, postal services and delivery services are increasingly relying on optical character recognition to recognize the addresses of the items to be delivered. As address recognition performance levels improve, it becomes increasingly difficult to make significant advances in performance with a single address recognition system. Consequently, two or more independent address recognition systems are applied and a method is required to arbitrate competing results from these systems to select on a case by case basis the result most likely to be accurate.

[0004] Present methods and systems for arbitrating competing results from two or more independent address recognition systems utilize a small subset of result characteristics. Present methods and systems use the level of “trust” associated with the address recognition engine that produced the result as the primary arbitration factor. In addition present methods and systems also use a coarse indication of an alphanumerical “postal code” (for example, ZIP Code in the U.S. Postal Service) matching against a limited set of results when from one or two other systems.

[0005] A method and system for arbitrating competing results from address recognition systems that are extendable to an arbitrary number of address recognition systems and which utilize a number of the characteristics of the results from the address recognition systems are needed.

SUMMARY OF THE INVENTION

[0006] A method and system to arbitrate between results obtained from observationally derived data by several procedures, where the results are items characteristic of the observationally derived data are disclosed. In the method, each procedure is given a ranking according to the confidence on the procedure. The results and characteristics derived from the results are used in a plurality of rules, where the rules are used to arbitrate between the results. The method generically comprises the steps of:

[0007] (a) classifying each result according to a level of accuracy at which it matches a datum from one of several databases,

[0008] (b) comparing each result corresponding to one observationally derived datum to results for that observationally derived datum obtained from other procedures,

[0009] (c) obtaining at least one indicator for each result, that indicator indicating a degree of success of the comparison of results,

[0010] (d) selecting a group of the rules according to an algorithm, where the rules utilize as inputs the results and characteristics derived from the results, the characteristics including the at least one indicator and the results of the comparative step,

[0011] (d) applying the selected rules according to the algorithm,

[0012] (e) obtaining, from applying the selected rules, arbitrated results.

[0013] The method can also include, but is not limited to, the steps of: comparing the results to a number of known incorrect results, and, discarding any of the results that matches any one of said known incorrect results, in which case, only the non-discarded results are arbitrated.

[0014] In one embodiment, observationally derived data is data derived from mail pieces, such as addresses, and the result is an alphanumerical code, such as a ZIP Code in the U.S. Postal Service. However, the invention need not be limited to delivery systems for addressed pieces and could be applied to any system where arbitration of results between several procedures is required.

[0015] For a better understanding of the present invention, together with other and further objects thereof, reference is made to the accompanying drawings and detailed description and its scope will be pointed out in the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

[0016] FIG. 1 is a flowchart of an embodiment of the method of this invention;

[0017] FIG. 2 is a graphical representation of an envelope as input to an address recognition system;

[0018] FIG. 3 is a flowchart of an embodiment of the method of this invention applied to arbitration of results of address recognition; and

[0019] FIG. 4 is a block diagram of an embodiment of a system of this invention used for arbitration of results of address recognition.

DETAILED DESCRIPTION

[0020] A method to arbitrate between results obtained from observationally derived data by several procedures, where the results are items characteristic of the observationally derived data is disclosed. An example of procedures that generate results from observationally derived data are address recognition systems utilizing optical character recognition recognizing the addresses of the items to be delivered in a postal system or a delivery system. The method of this invention is extendable to an arbitrary number of procedures and utilizes a number of the characteristics of the results from the procedures.

[0021] A flowchart of an embodiment of the method of this invention is shown in FIG. 1. Results from several procedures 10 are compared to data from one of several databases 40 and classified according to a level of accuracy at which it matches a datum from one of the several databases 40 (step 70, FIG. 1).

[0022] The method can include a pre-processing series of steps in which the results 10 are compared to known incorrect results 30 (steps 50 and 60, FIG. 1). Some of procedures can generate known incorrect results as part of the normal course of operation of the procedure. For example, a procedure can have access to several databases of a different precision. Results from such a procedure that are inconsistent when compared to data bases of a different precision are considered incorrect results. In the first of the pre-processing steps, the results 10 are compared to known incorrect results 30 (step 50, FIG. 1). Any of the results 10 that matches one of the known incorrect results 30 is discarded (step 60, FIG. 1). The non discarded results are then compared to data from one of several databases 40 and classified according to a level of accuracy at which it matches a datum from one of the several databases 40 (step 70, FIG. 1).

[0023] Each result 10 corresponding to one observationally derived datum is compared to results 10 for the same observationally derived datum obtained from other ones of the procedures (step 80, FIG. 1). At least one indicator, indicating a degree of success of the comparison of results for each result corresponding to the same observationally derived datum, is assigned to each result 10 (step 90, FIG. 1).

[0024] Each procedure has a given ranking 120 according to the confidence on that procedure. An indicator of the ranking of confidence of the procedure that produced the result (hereafter called a confidence value) is assigned to each result (step 130, FIG. 1).

[0025] A data structure (not shown) can be constructed including the results, an identifier for each of the procedures that produced the results, the ranking according to confidence for each of the procedures, the confidence value for each result, and, at least one indicator for each result indicating the degree of success of the comparison of results for each result.

[0026] A set of rules 20, aimed at hierarchically obtaining a maximum number of successful results, is utilized in the arbitration process. The rules codify known practices and desired procedures for obtaining successful results or eliminating unsuccessful results. The rules utilize as inputs the results 10 and characteristics derived from the results, the characteristics including the at least one indicator and the results of the comparison of results (performed in step 80, FIG. 1). The results 10 (not including the discarded ones if pre-processing was performed), the at least one indicator, indicating the degree of success of the comparison of results for each result, the confidence value for each result 10 and selected ones of the rules 20 are inputs to an algorithm applying the selected rules to arbitrate between results corresponding to the same observationally derived datum but obtained by different procedures (step 100, FIG. 1). The outcome is the determination, from the arbitration, of a procedure, from which results most likely to be accurate are obtained (step 110, FIG. 1). Another indicator indicative of the outcome of applying the selected rules is appended to each result (step 120, FIG. 1). An embodiment of the method as applied to arbitrating competing results from two or more independent address recognition systems used in postal systems or package delivery systems is described below.

[0027] Arbitration of Results of Address Recognition Systems

[0028] When two or more independent address recognition systems utilizing optical character recognition are used in postal services or package delivery services, the result used for arbitration is an alphanumerical code characterizing each address. (In the U.S. Postal Service, the alphanumerical code is a ZIP Code, a five, nine or eleven digit code, the first five of which are most commonly used.) In such systems, an addressed piece, such as envelope 135 shown in FIG. 2, is scanned by scanning means and the scanned output is provided as input to the address recognition systems utilizing optical character recognition. The optical character recognition (OCR) modules provide data on the delivery address 140, the alphanumerical code 150 (herein after referred to as the ZIP Code) and the return address 160 as well as other possible items on the addressed piece 135, such as a bar code (not shown).

[0029] A flowchart of an embodiment of the method of this invention applied to arbitration of results of address recognition systems is shown in FIG. 3. Result ZIPs from several procedures 210 are compared to data from one of several databases 240 and classified according to a level of accuracy at which it matches a datum from one of the several databases 40 (step 270, FIG. 3). Step 270 is hereinafter referred to as classifying as to sort depth.

[0030] The method can include a pre-processing series of steps in which the result ZIPs 210 are compared to known “Bad ZIPs” 230 (steps 250 and 260, FIG. 3). Some address recognition systems are able to assert as part of their input to the arbitration process “bad” ZIP Codes developed internally as a possible delivery address, but rejected for specific attributes that indicate an error situation. For example, an address recognition system may have access to multiple address databases 240 of different precision. One address database (low precision) might represent a range between starting and ending address numbers on a street where a second database (high precision) lists the addresses only for existing dwellings on the street. An address recognition system with access to both databases can provide as input to the arbitration process ZIPS which are “bad” in the sense that they are inconsistent with the high precision database, but are likely outcomes from engines that have access only to a low precision database. In the first of the pre-processing steps, the results 10 are compared to known incorrect results 30 (step 50, FIG. 1). Any of the result ZIPs 210 that matches one of the known “Bad” ZIPs 230 is discarded (step 260, FIG. 3). The non discarded ZIPs are then compared to data from one of several databases 240 and classified according to “sort depth” (step 270, FIG. 3).

[0031] Each result ZIP 210 corresponding to one observationally derived datum is compared to results 210 for the same observationally derived datum obtained from other ones of the procedures (step 280, FIG. 3). The arbitration method considers the quantity and extent of matches and mismatches between ZIP results from different address recognition procedures (also referred to as systems). For USA ZIP Codes, a distinction is made in the arbitration process between mismatches in four sub-fields of the ZIP Code:—High order 3 digits (which identify a large region, sometimes incorporating multiple cities or towns)—High order 5 digits (typically unique within a single city) High order 9 digits (typically unique to a carrier route)—11 digits (identifying a specific delivery point). As an example of arbitration process sensitivity to the extent of mismatch, the process considers mismatches in the high order 3 digits as a possible indication that one of the engines has asserted a return address result instead of the intended delivery address. At least one indicator, indicating a degree of success of the comparison of results for each result corresponding to the same observationally derived datum, is assigned to each result ZIP 210 (step 290, FIG. 3).

[0032] Implementations of such indicators are a 9-Match Count and a 5-Match Count defined below. The 9-Match Count is obtained by comparing 9 digits of each Result ZIP 210 to result ZIPs 210 from other processes, including ZIPS of lower or finest depth. The 9-Match is incremented for each match against a ZIP created by a different database matching technique (hereinafter referred to as a DLU). Each different DLU is counted only once. The 5-Match Count is defined by comparing 5 digits of each Result ZIP 210 to Result ZIPs 210 from other processes. Increment the 5-Match for each match against a ZIP created by a different Address Block Location (ABL) algorithm.

[0033] Each procedure has a given ranking 120 according to the confidence on that procedure. An indicator of the ranking of confidence of the procedure that produced the result (hereafter called a confidence value) is assigned to each result ZIP (step 330, FIG. 3).

[0034] A data structure can be constructed including the result ZIPs 210, an identifier for each of the procedures that produced the results, the ranking 120 according to confidence for each of the procedures, the 5-Match Count, the 9-Match Count, and the sort depth. An implementation of the data structure is shown in Table 1 below.

TABLE 1
Result structure
							Sort
							Depth
							(Finest,
							and	5-	9-
Result	Finalized				DLU	ABL	enumerated	Match	Match
ZIP	ZIP	Source	Class	Subclass	type	type	types)	Count	Count
	DS	0	0	RCR	RCR
	Char—	0	1	RCR	MLOCR
	M
	Char—	0	2	RCR	RCP
	C
	B	0	3	B	B
	Gray
	CSAR	1	0	USPS	RCP or
				or B	MLOCR
	D	2	0	B	MLOCR
	M	2	1	USPS	MLOCR
	C	2	2	USPS	RCP

[0035] The Class and Subclass items in the Result structure of Table 1 are indicators of the quality or precision of the results expected (or confidence for each of the procedures). Class 0 indicates the highest precision or highest confidence results; Class 2 indicates a lower precision or lower confidence result. That is, the lower the class number the higher the accuracy (confidence) of the result. Subclasses serve to provide a finer ranking of results according to precision (confidence). Subclass 0 indicates the highest precision (confidence) among the subclasses; Subclass 2 indicates a lower precision (confidence) result among the subclasses. Thus, class 0 and subclass 0 would produce the ranking of highest confidence. Class 2, subclass 2 indicates a lower precision (confidence) result (the lowest shown in Table 1). The indicator of the ranking of confidence of the procedure that produced the result, the confidence value, can be obtained from the class, subclass values.

[0036] A set of rules 20, aimed at hierarchically obtaining a maximum number of successful results, is utilized in the arbitration process. The rules codify known practices and desired procedures for obtaining successful results or eliminating unsuccessful results. The result ZIPs 210 (not including the discarded ones if pre-processing was performed), the at least one indicator, indicating the degree of success of the comparison of results for each result, the confidence value for each result 310 and selected ones of the rules 20 are inputs to an algorithm applying the selected rules to arbitrate between results corresponding to the same observationally derived datum but obtained by different procedures (step 300, FIG. 3). The outcome is the determination, from the arbitration, of a procedure, from which results most likely to be accurate are obtained (step 310, FIG. 3). A second indicator, a Delivery Print Code (DPC), indicative of the outcome of applying the selected rules is appended to each result (step 320, FIG. 3). As an outcome of applying the selected rules, a ZIP Code is selected corresponding to the result selected from arbitration among procedures. This ZIP Code provides a major datum for address recognition.

[0037] One embodiment of the rules is given below.

[0038] Return-to-Sender/Forwarding Detection:

[0039] 1) System Reject if return-to-sender/forwarding case has been found from any source.

[0040] This is an implementation of a rule that comprises detecting specific information in observationally derived data. Other implementations include detection by one or more address recognition systems that the mail piece falls into a specific class such as Business Reply.

[0041] The “System Reject” outcome indicates that there is a very high likelihood that a ZIP result asserted by any of the procedures is very likely incorrect. For example, if one of the procedures detected with certainty that the mail piece has been marked as “Return to Sender”, it would be unadvisable to accept ZIP results from the other procedures. If those other procedures are unaware of the “Return to Sender” indication, the destination address could be erroneously asserted (rather than, correctly, the return address) as a ZIP to be arbitrated. Once asserted, the “System Reject” outcome is retained as a piece of data in the electronic record that is associated with the mail piece, throughout the remaining automated mail processing steps.

[0042] Some address assignment systems have an on-line (real-time) phase and an off-line phase. Address resolution for the on-line phase must typically occur within seconds, while minutes or hours are allocated for off-line processing. Longer-running algorithms may be applied during the off-line phase to resolve difficult cases. The arbitration process for on-line processing will assert a “System Reject” outcome to prevent an erroneous result from occurring during off-line processing. The “System Reject” status can result in the mail piece receiving special handling (human intervention) during the off-line phase.

[0043] Mismatch in the First Three Digits of a ZIP (Known as SCF) Cases:

[0044] 2) System Reject if there is an SCF mismatch among the combined set of finalized Class 0 and Class 1 ZIPs.

[0045] 3) System Reject if there is an SCF mismatch among the combined set of finalized and partial Class 0 and Class 1 ZIPs.

[0046] 4) System Reject if there is an SCF mismatch among the combined set of finalized Class 0, Class 1, and Class 2 ZIPs.

[0047] 5) System Reject if there is an SCF mismatch among the combined set of finalized and partial Class 0, Class 1, and Class 2 ZIPS.

[0048] Conflicting Foreign:

[0049] 6) Reject if there is a zone mismatch (assignments to different countries) among the combined set of finalized foreign Class 0, Class 1, and Class 2 ZIPs.

[0050] The above five rules (2 through 6) are implementations of rules that detect a lack of congruency of a group of digits in each one of said alphanumerical codes obtained from specific observationally derived data by several systems.

[0051] Supported Finest Depth ZIPS:

[0052] 7) Select as arbitrated result the highest ranked (by Class, Subclass) Finest Depth ZIP with a 9-Match >0.

[0053] The above is an implementation of a rule that comprises ignoring, when more than one result is obtained corresponding to one observationally derived datum by several procedures, the results corresponding to that one observationally derived datum and obtained from the procedures having a lower confidence ranking.

[0054] 8) Select as arbitrated result the highest ranked (by Class, Subclass) Finest Depth ZIP (RCR). This is a variation of rule 7.

[0055] 5-Digit Match Among Finalized Results:

[0056] 9) If there is no 5-digit mismatch among finalized results from Class 0, Class 1, and Class 2:

[0057] Order the ZIPs within the combined set of Class 0 and Class 1 by sort depth, then 9-Match, then Class. Select as arbitrated result the highest ranked finalized result.

[0058] 10) If there is no 5-digit mismatch among finalized results from Class 0, Class 1, and Class 2:

[0059] Order the ZIPs within Class 0 by sort depth, then 9-Match, then Subclass. Select as arbitrated result the highest ranked finalized result.

[0060] 11) If there is no 5-digit mismatch among finalized results from Class 0, Class 1, and Class 2:

[0061] Order the ZIPs within the combined set of Class 0 and Class 1 by 9-Match, then Class, then sort depth. Select as arbitrated result the highest ranked finalized result.

[0062] The above three rules are implementations of rules that comprise detecting congruency of a group of digits in each one of the alphanumerical codes obtained from specific observationally derived data by several procedures.

[0063] 12) If there is no 5-digit mismatch among finalized results from Class 0, Class 1, and Class 2:

[0064] Order the ZIPs within Class 0 (and any from Classes 1 and 2 that are inherently finest depth) by sort depth, then 9-Match, then Subclass. Select as arbitrated result the highest ranked finalized result. This is a variation of rule 9.

[0065] 5-Digit Mismatch Among Finalized Results:

[0066] 13) If there is a 5-digit mismatch among finalized results from Class 0, Class 1, and Class 2:

[0067] Order the ZIPs within the combined set of Class 0 and Class 1 by sort depth, then 5-match, then 9-Match, then Class. Select as arbitrated result the highest ranked finalized result.

[0068] 14) If there is a 5-digit mismatch among finalized results from Class 0, Class 1, and Class 2:

[0069] Order the ZIPs within the combined set of Class 0 and Class 1 by 5-Match, then 9-match, then Class, then sort depth. Select as arbitrated result the highest ranked finalized result.

[0070] 15) If there is a 5-digit mismatch among finalized results from Class 0, Class 1, and Class 2:

[0071] Order the ZIPs within the combined set of Class 0 and Class 1 by 5-match, then 9-Match, then Class, then sort depth. Select as arbitrated result the highest ranked finalized result with a 9-Match greater than zero. Otherwise reject.

[0072] 16) If there is a 5-digit mismatch among finalized results from Class 0, Class 1, and Class 2:

[0073] Order the ZIPs within Class 0 by 9-Match, then 5-match, then Subclass, then sort depth. Select as arbitrated result the highest ranked finalized result with a 9-Match greater than zero. Otherwise reject.

[0074] 17) If there is a 5-digit mismatch among finalized results from Class 0, Class 1, and Class 2:

[0075] Order the ZIPs within Class 0 (and any from Classes 1 and 2 that are inherently finest depth) by 9-Match, then 5-match, then Subclass, then sort depth. Select as arbitrated result the highest ranked finalized result with a 9-Match greater than zero. Otherwise reject. This is a variation of rule 14.

[0076] The above five rules are also implementations of rules that comprise detecting congruency of a group of digits in each one of the alphanumerical codes obtained from specific observationally derived data by several procedures.

[0077] By the definition of the 9 Match Count and the 5 Match Count, the rules that utilize those Counts are implementations of rules that comprise utilizing knowledge of details of each of the procedures to determine a confidence level to each comparison of each result corresponding to one observationally derived datum to another result for the one observationally derived datum obtained from another one of the procedures.

[0078] No Finalized Results from RCR:

[0079] 18) Select Current System Arbitrated Result (after Bad ZIP filtering) as arbitrated result, if there are no Class 0 finalized ZIPs.

[0080] 19) Select Current System Arbitrated Result as arbitrated result, if it's a non-finalized 5 digit ZIP (partial result) and there are no Class 0 finalized ZIPs. Otherwise reject.

[0081] 20) Select Current System Arbitrated Result as arbitrated result, if it's a non-finalized 5 digit ZIP (partial result) and there are no Class 0 finalized ZIPs. This is a variation of rule 16.

[0082] ID Tag:

[0083] 21) If there are no five-digit finalizations among the Class 0 results, and if the ID-tag time stamp is more than 76 hours older than a control time stamp, then reject.

[0084] The above rule is an implementations of a rule that comprises a relationship between a time period, that time period comprising the difference between when a mail piece was sent and when the observationally derived data was obtained, and a probability of inaccuracy of the results obtained from the observationally derived data corresponding to that mail piece.

[0085] Selected ones of the above rules are applied together according to a decision algorithm in order arbitrate between the results of several independent address recognition systems applied to the same data. The implementations given below operate on an ordered sequence of selected ones of the above rules.

[0086] Algorithm A1

[0087] Rule Sequence: 1, 21, 18, 7, 3, 6, 9, 13

[0088] Comments:

[0089] Picks finest depth result, despite conflicts at 5-digit level.

[0090] Algorithm A2

[0091] Rule Sequence: 1, 21, 18, 7, 3, 6, 9, 13

[0092] Comments:

[0093] Do not perform bad ZIP filtering.

[0094] Picks finest depth result, despite conflicts at 5-digit level.

[0095] Algorithm A3

[0096] Rule Sequence: 1, 21, 18, 7, 3, 6, 9, 13

[0097] Comments:

[0098] Do not perform bad ZIP filtering.

[0099] 9-Match count is incremented only if the ZIPs match to 11 digits.

[0100] Picks finest depth result, despite conflicts at 5-digit level.

[0101] Algorithm B1 (Nominal Error Rate)

[0102] Rule Sequence: 1, 21, 18, 7, 3, 6, 9, 14

[0103] Comments:

[0104] Should perform closest to the competitive test arbitration; differences:

[0105] SCF mismatch includes conflicts with character-based results

[0106] Algorithm B2 (Nominal Error Rate)

[0107] Rule Sequence: 1, 21, 18, 7, 3, 6, 9, 14

[0108] Comments:

[0109] Do not perform bad ZIP filtering.

[0110] Should perform closest to the competitive test arbitration; differences:

[0111] SCF mismatch includes conflicts with character-based results

[0112] Algorithm B3 (Nominal Error Rate)

[0113] Rule Sequence: 1, 21, 18, 7, 3, 6, 9, 14

[0114] Comments:

[0115] Do not perform bad ZIP filtering.

[0116] 9-Match count is incremented only if the ZIPs match to 11 digits.

[0117] Should perform closest to the competitive test arbitration; differences:

[0118] SCF mismatch includes conflicts with character-based results

[0119] Algorithm B4 (Nominal Error Rate)

[0120] Rule Sequence: 1, 21, 18, 7, 3, 6, 9, 14

[0121] Comments:

[0122] Do not perform bad ZIP filtering.

[0123] 9-Match count is incremented only if the ZIPs match to 11 digits and both the ABLs and DLUs are from different sources.

[0124] Should perform closest to the competitive test arbitration; differences:

[0125] SCF mismatch includes conflicts with character-based results

[0126] Algorithm C1 (Lower SCF Error Rate)

[0127] Rule Sequence: 1, 21, 18, 3, 6, 7, 9, 14

[0128] Comments:

[0129] Slightly lower SCF error rate due to applying SCF mismatch early in sequence, over-riding all assignments

[0130] Algorithm D1 (Even Lower SCF Error Rate)

[0131] Rule Sequence: 1, 21, 18, 5, 6, 7, 9, 14

[0132] Comments:

[0133] Lower SCF error rate due to applying SCF mismatch against results from all individual engines (not just against arbitrated current system results)

[0134] Algorithm E1 (Lower Overall Error Rate)

[0135] Rule Sequence: 1, 21, 18, 5, 6, 7, 11, 15

[0136] Comments:

[0137] Lower error rate due to conservative approach for 5-digit mismatch cases—only asserting an answer if different DLUs arrive at a matching 9/11digit finalization.

[0138] Algorithm F1 (Very Low Overall Error Rate)

[0139] Rule Sequence: 1, 21, 19, 5, 6, 7, 10, 16

[0140] Algorithm F2 (Very Low Overall Error Rate)

[0141] Rule Sequence: 1, 21, 20, 5, 6, 7, 12, 17

[0142] The rules and algorithms given above are not an exhaustive list. Other rules and algorithms can be applied and other characteristics of the results can be utilized as further inputs to the rules. For example, counts for mismatches of varying significance (5 digit, 9 digit) between competing results could be utilized as inputs to the rules or the algorithms. Similarly, a second confidence value associated with each result, asserted by the procedure that produces the ZIP result, could also be used as input to the rules or the algorithms. The second confidence value would be asserted by the procedure based on information regarding the observationally derived data (perhaps based on knowing that OCR results themselves were poor for that mail piece image, or that words in the address are apparently misspelled). The second confidence value would provide another input that could be utilized by an algorithm to determine a procedure, from which results most likely to be accurate are obtained.

[0143] The above algorithms implement a sequential application of selected ones of the above rules in order to codify known operational or observed performance. A known method for achieving the same is through the application of the techniques of artificial intelligence. The algorithms and rules are already in a form akin to an inference engine and can be cast into a rule based system for decisions or an “Expert System” (see, for example, P. Harmon, D. King, Expert Systems—Artificial Intelligence in Business, ISBN 0-471-80824-5, 1985). Following more recent trends in artificial intelligence, the algorithms can be cast in the form of a neural network (see, for example, S. K. Rogers, M. Kabrisky, An Introduction to Biological and Artificial Neural Networks for Pattern Recognition, ISBN 0-8194-0534-5, 1991) or a genetic algorithm (also known as an evolutionary algorithm). In either the neural network or the genetic algorithm, the rules can be generated by the neural network or the genetic algorithm. It should be apparent that, as other methods are developed in artificial intelligence or machine aided cognition, the algorithms can also be cast in the form of those methods.

[0144] The above method for arbitration between results obtained by more than one method for matching observationally derived data to data from a number of databases can be implemented as a computer readable code embodied in a computer usable or readable medium.

[0145] A block diagram of an embodiment of a system 400 of this invention used for arbitration of results of address recognition is shown in FIG. 4. The scanned output is provided as digitized input 420 to the address recognition system. A processor 410 operates on the input 420 and, utilizing the OCR systems implemented as computer readable code stored in a memory 460 and the database 425 generates address results. The processor 410, then, executes the arbitration method implemented as a computer readable code embodied in the memory 460. Other Memory 440 is used for other system functions (for example, control of scanning and transport, data and event logging, user interface, and other “housekeeping” functions) and could be implemented by means of any computer readable media. An interface to electromechanical systems 470 allows communication and control of scanning and transport functions. Such an interface could be specific to the actual design of the scanning and transport functions but such interfaces are well known in the art.

[0146] In general, the techniques described above may be implemented, for example, in hardware, software, firmware, or any combination thereof. The techniques described above may be implemented in one or more computer programs executing on a programmable computer including a processor, a storage medium readable by the processor (including, for example, volatile and non-volatile memory and/or storage elements), at least one input device, and at least one output device. Program code may be applied to data entered using the input device to perform the functions described and to generate output information. The output information may be applied to one or more output devices.

[0147] Elements and components described herein may be further divided into additional components or joined together to form fewer components for performing the same functions.

[0148] Each computer program within the scope of the claims below may be implemented in any programming language, such as assembly language, machine language, a high-level procedural programming language, or an object-oriented programming language. The programming language may be a compiled or interpreted programming language.

[0149] Each computer program may be implemented in a computer program product tangibly embodied in a computer-readable storage device for execution by a computer processor. Method steps of the invention may be performed by a computer processor executing a program tangibly embodied on a computer-readable medium to perform functions of the invention by operating on input and generating output.

[0150] Common forms of computer-readable or usable media include, for example, a floppy disk, a flexible disk, hard disk, magnetic tape, or any other magnetic medium, a CDROM, any other optical medium, punched cards, paper tape, any other physical medium with patterns of holes, a RAM, a PROM, and EPROM, a FLASH-EPROM, any other memory chip or cartridge, a carrier wave, or any other medium from which a computer can read.

[0151] Although the invention has been described with respect to various embodiments, it should be realized this invention is also capable of a wide variety of further and other embodiments within the spirit and scope of the appended claims.

Claims

1. A method to arbitrate between results to effectuate a desired result, said results to be arbitrated obtained by a plurality of procedures from observationally derived data, said method comprising the steps of: a. ranking each of the procedures according to confidence on each of said procedures; b. classifying each of the results according to a level of accuracy at which each of said results matches a datum from a database of a plurality of databases; c. comparing each of the results to each other; d. generating rules, said rules utilizing as inputs said results and information derived from said results; e. selecting at least one of said rules; f. applying said at least one of said rules according to an algorithm; g. obtaining, from the application said at least one of said rules, said desired result.

2. The method of claim 1 further comprising the step of: appending an indicator to each result, said indicator indicating an outcome of applying said at least one rule.

3. The method of claim 1 further comprising the steps of: comparing the results to a plurality of known incorrect results; discarding any of said results that matches any one of said known incorrect results; and, wherein steps (a) through (g) are applied only to the said results that are not discarded.

4. The method of claim 1 further comprising the steps of: obtaining at least one other indicator for each result, said at least one other indicator indicating a degree of success of said comparative step for each observationally derived datum.

5. The method of claim 4 further comprising the steps of: creating a data structure, said data structure comprising: said results; said desired result obtained by applying said at least one of said rules; one further other indicator indicating said ranking according to confidence for each of said results; said at least one other indicator for each of the results.

6. The method of claim 1 further comprising the step of: assigning to each of said results yet another indicator of said ranking of confidence of the procedure that produced said result.

7. The method of claim 1 wherein said observationally derived data comprises data derived from addressed pieces and wherein said result comprises an alphanumerical code.

8. The method of claim 7 further comprising the step of: obtaining at least one indicator for one of said alphanumerical codes, said one of said alphanumerical codes resulting from applying one of said procedures to specific observationally derived data, said at least one indicator indicating the number of alphanumerical codes resulting from applying other ones of said plurality of procedures to said specific observationally derived data that are congruent to said one of said alphanumerical codes for a given number digits of said one of said alphanumerical codes.

9. The method of claim 1 wherein said rules comprise detecting specific information in observationally derived data.

10. The method of claim 7 wherein said rules comprise detecting return or forwarding information in said data derived from addressed pieces.

11. The method of claim 7 wherein said rules comprise detecting a lack of congruency of a group of digits in each one of said alphanumerical codes obtained from specific observationally derived data by several ones of said plurality of procedures.

12. The method of claim 7 wherein said rules comprise detecting congruency of a group of digits in each one of said alphanumerical codes obtained from specific observationally derived data by several ones of said plurality of procedures.

13. The method of claim 7 wherein said rules comprise ignoring, when more than one result is obtained corresponding to one observationally derived datum by applying a plurality of procedures, the results corresponding to said one observationally derived datum and obtained from other ones of said plurality of procedures having a lower ranking.

14. The method of claim 7 wherein said rules comprise a relationship between a time period, said time period comprising the difference between when an addressed piece was sent and when said observationally derived data was obtained, and a probability of inaccuracy of said results obtained from the observationally derived data corresponding to said addressed piece.

15. The method of claim 7 wherein said rules comprise utilizing knowledge of details of each of said plurality of procedures to determine a confidence level to each said comparison of each result corresponding to one observationally derived datum to another result for said one observationally derived datum obtained from another one of said plurality of procedures.

16. A computer program product comprising: a computer usable medium having computer readable code embodied therein for arbitrating between results to effectuate a desired result, said results obtained from observationally derived data by a plurality of procedures, said code causing a computer system to: (a) rank each of the procedures according to confidence on each said procedure; (b) classify each of the results according to a level of accuracy at which each of the results matches a datum from a database of a pluralities of databases; (c) compare each of the results to each other; (d) select at least one rule, said at least one rule utilizing as inputs said results and information derived from said results; (e) apply said at least one rule according to an algorithm; (f) obtain, from applying said selected rules, said desired result.

17. The computer program product of claim 16 where the computer readable code further causes a computer system to: compare the results to a plurality of known incorrect results; discard any of said results that matches any one of said known incorrect results; and, wherein the computer system applies steps (a) through (f) to the non-discarded results.

18. The computer program product of claim 16 where the computer readable code further causes a computer system to: append an indicator to each result, said indicator indicating an outcome of applying said at least one rule.

19. The computer program product of claim 16 where the computer readable code further causes a computer system to: obtain at least one other indicator for each result, said at least one other indicator indicating a degree of success of said comparative step.

20. The computer program product of claim 19 where the computer readable code further causes a computer system to: create a data structure, said data structure comprising: said results; said desired result obtained by applying said selected at least one rule; one further other indicator indicating said ranking according to confidence for each of said results; said at least one other indicator for each result.

21. The computer program product of claim 16 wherein said observationally derived data comprises data derived from addressed pieces and wherein said result comprises an alphanumerical code.

22. The computer program product of claim 16 where the computer readable code further causes a computer system to: generate said at least one rule.

23. The computer program product of claim 21 where the computer readable code further causes a computer system to: obtain at least one indicator for one of said alphanumerical codes, said one of said alphanumerical codes resulting from applying one of said procedures to specific observationally derived data, said at least one indicator indicating the number of alphanumerical codes resulting from applying other ones of said plurality of procedures to said specific observationally derived data that are congruent to said one of said alphanumerical codes for a given number digits of said one of said alphanumerical codes.

24. A system including arbitration between results to effectuate a desired result, said results obtained by a plurality of procedures from observationally derived data, the result being items characteristic of the observationally derived data, said system comprising: at least one processor; at least one computer readable memory having instructions embodied therein, said instructions causing said at least one processor to: rank each of the procedures according to confidence on each said procedure; classify each of the results according to a level of accuracy at which it matches a datum from a database of a plurality of databases; compare each of the results to each other; select at least one rule, said at least one rule utilizing as inputs said results and characteristics derived from said results; apply said at least one rule according to an algorithm; obtain, from applying said at least one rule, said desired result.

25. The system of claim 24 wherein the instructions embodied in the at least one computer readable memory further cause said at least one processor to: append an indicator to each result, said indicator indicating an outcome of applying said at least one rule.

26. The system of claim 24 wherein the instructions embodied in the at least one computer readable memory further cause said at least one processor to: compare the results to a plurality of known incorrect results; discard any of said results that matches any one of said known incorrect results.

27. The system of claim 24 wherein the instructions embodied in the at least one computer readable memory further cause said at least one processor to: obtain at least one other indicator for each of the results, said at least one other indicator indicating a degree of success of said comparison of results for each of the results.

28. The system of claim 27 wherein the instructions embodied in the at least one computer readable memory further cause said at least one processor to: create a data structure, said data structure comprising: said results; said desired result obtained by applying said selected at least one rule; one further other indicator indicating said ranking according to confidence for each of said results; said at least one other indicator for each of the results.

29. The system of claim 24 wherein said observationally derived data comprises data derived from addressed pieces and wherein said results comprise alphanumerical codes.

30. The system of claim 29 wherein the instructions embodied in the at least one computer readable memory further cause said at least one processor to: obtain at least one indicator for one of said alphanumerical codes, said one of said alphanumerical codes resulting from applying one of said procedures to specific observationally derived data, said at least one indicator indicating the number of alphanumerical codes resulting from applying other ones of said plurality of procedures to said specific observationally derived data that are congruent to said one of said alphanumerical codes for a given number digits of said one of said alphanumerical codes.

31. The system of claim 24 wherein the instructions embodied in the at least one computer readable memory further cause said at least one processor to: generate said at least one rule.

32. The method of claim 1 wherein said algorithm is implemented using techniques of artificial intelligence.

33. The computer program product of claim 16 wherein said algorithm is implemented using techniques of artificial intelligence.

34. The computer program product of claim 22 wherein said algorithm is implemented using techniques of artificial intelligence.

35. The system of claim 24 wherein said algorithm is implemented using techniques of artificial intelligence.

36. The system of claim 31 wherein said algorithm is implemented using techniques of artificial intelligence.