Automatic linking of requirements using natural language processing

Patent metadata

Field	Value
Publication number	US-9342489-B2
Application number	US-201414243972-A
Country	US
Kind code	B2
Filing date	Apr 3, 2014
Priority date	Apr 3, 2014
Publication date	May 17, 2016
Grant date	May 17, 2016

Abstract

A method of automatic identifying linking relationships of requirements in a plurality of requirement documents. Terms in the plurality of requirement documents are identified. A part-of-speech tag is assigned to each term. Each identified term is selected as a focal term. Co-occurring terms within a predetermined distance of the selected focal term are determined. A linking relationship probability is calculated for each co-occurring term associated with the selected focal term. The selected focal terms and associated co-occurring terms between the plurality of requirement documents are compared. A degree of linking relationship is identified between two requirements as a function of a comparison between selected focal terms and the associated co-occurring terms between the plurality of requirement documents. An analysis report identifying the degree of linking relationships between two respective requirements is output.

First claim

What is claimed is:

1. A method of identifying linking relationships of requirements in a plurality of requirement documents comprising the steps of:
(a) identifying terms in the plurality of requirement documents;
(b) assigning a part-of-speech tag to each term, the part-of-speech tag indicating whether the term is a part term, symptom term, action term or failure mode term;
(c) selecting each identified term as a focal term and respectively determining co-occurring terms within a predetermined distance of the selected focal term;
(d) calculating a linking relationship probability for each co-occurring term associated with the selected focal term;
(e) repeating steps (a)-(d) for each selected focal term in the plurality of requirement documents;
(f) comparing the selected focal terms and associated co-occurring terms between the plurality of requirement documents;
(g) identifying a degree of linking relationship between two requirements as a function of a comparison between selected focal terms and the associated co-occurring terms between the plurality of requirement documents; and
(h) outputting an analysis report identifying the degree of linking relationships between two respective requirements.

2. The method of claim 1 further comprising the steps of
identifying whether the term is a member of a phrase in response to identifying terms in the plurality of the requirement documents; and
grouping the phrases having a same number of terms.

3. The method of claim 2 further comprising the steps of:
identifying the critical phrases within the identified phrases; and
disregarding non-critical phrases.

4. The method of claim 3 wherein the critical phrases are determined by the formula:
( C n-gram ) i,j =C i,j *idf Ci
where C i,j relates to an appearance of ith phrase in a jth requirement document, and idf Ci is a weighting factor.

5. The method of claim 4 wherein the wherein C i,j is represented by the following formula:

C

i
,
j

=

n

i
,
j

∑
k

⁢

⁢

n

k
,
j

where n i,j is the number of occurrences of a given n-gram, C i is the appearance of ith phrase in a requirement document C j , and the denominator is the sum of number of occurrence of all n-grams in C j .

6. The method of claim 4 wherein idf Ci is represented by the following formula:

idf
Ci

=

log
⁢


V




{

v
⁢

:

⁢

⁢

C
i

∈
v

}



where |V| is a total number of requirement documents in a corpus, and {v: C i εv} is number of requirement documents with the records of C i .

7. The method of claim 4 wherein the phrase is identified as a critical phrase if the probability determined for the phrase is above a predetermined percentage of the probabilities determined for all phrases identified.

8. The method of claim 7 wherein the predetermined percentage is 85%.

9. The method of claim 1 wherein assigning a part-of-speech tag associated with a term includes labeling a term with one of a noun, verb, adjective, and adverb, and wherein the part-of-speech tag is used to determine whether the term is a part, symptom, action, or failure mode term.

10. The method of claim 1 further comprising a step of calculating a probability of the term being associated with the respective part-of-speech tag in response to the respective term being assigned a part-of-speech tag.

11. The method of claim 10 wherein the probabilities are calculated using a maximum likelihood estimation as follows:

P
⁡

(

pos
-

tag
m

|

pos
-

tag

m
-
1

)

=

f
⁡

(

pos
-

tag

m
-
1

,

pos
-

tag
m

)

f
⁡

(

pos
-

tag

m
-
1

)

P
⁡

(

n
-

Assignees

• Gm Global Tech Operations Inc

Inventors

• Sethu Ramesh
• Peranandam Prakash M
• Rajpathak Dnyanesh
• De Soumen

Classifications

• G06F40/194 Primary

  Calculation of difference between files · CPC title

• G06F40/284

  Lexical analysis, e.g. tokenisation or collocates · CPC title

• G06F40/289

  Phrasal analysis, e.g. finite state techniques or chunking · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06F17/277

  Physics · mapped topic