DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2 are rejected under 35 U.S.C. 103 as being unpatentable over Hazrati et al, NPL ref: Evaluation of Unknown Groundwater Contaminant Sources Characterization Efficiency under Hydrogeologic Uncertainty in an Experimental Aquifer Site by Utilizing Surrogate Models Shahrbanoo Hazrati-Yadkoori1*, Bithin Datta , hereinafter Hazrati , in view of Yao et al (Ref CN 112241844 A) hereinafter referenced as Yao, in view of NPL Ref: Wani Tamas et al, Hybridization of Air quality forecasting Models Using Machine Learning and Clustering, Aerosol and Air Quality Research, 16:405-416, 2016, hereinafter Tamas, and further in view of NPL Ref: Self- and Super-organizing Maps in R: The kohonen Package, Wehrens et al hereinafter Wehrens.
Claim 1, Hazrati teaches “a method for identifying groundwater pollution source”; (Hazrati, [Abstract] ,“Characterization of unknown groundwater contaminant sources.”) and “chemical index concentration data, pollutant concentration data”; (Hazrati, page 12, 1st paragraph, “To analyze the background chemical concentrations of tested elements, 88 groundwater samples were collected”) and teaches “longitude and latitude coordinates”;(Hazrati, page 9, Figure 3, “site location” reads on longitude and latitude coordinate).
Note: Hazrati teaches self-organizing maps SOM, which is the same approach of this application in light of the specification [0017]. Tamas also teaches self-organizing maps k-SOM.
Hazrati does not teach “surface water system data, and enterprise type data;”
Yao teaches “surface water system data, and enterprise type data;” (Yao, page 3, 11th paragraph, “risk point calculation model of point source type risk is S1 = G1 (a, b, c, d, e, f1,g1); “f1 represents the enterprise information”, “b represents the water system information”).
Therefore, it would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine the teaching of Hazrati groundwater and concentrated pollution chemical data and location or source with Yao to calculate and identify water surface risk with enterprise type information wherein the enterprise information is well-known in the art and would yield the predictable results yet with higher feasibility. (ref KSR Int’l Co. v. Teleflex Inc.)
Hazrati, Yao does not teach ”Euclidean distance and clustering distance between sample data and a corresponding output neuron weight vector”
Tamas teaches “calculating a Euclidean distance and clustering distance between sample data and a corresponding output neuron weight vector;” (page 415, col 1 , para 2, “Euclidean distance with Ward criterion (hMLP), and k-means clustering coupled with Self Organisation Map (kMLP)”, hMLP (hierarchical Clustering) reads on output neuron weight vector; page 408 col 2, 2nd paragraph, “Hierarchical clustering is an iterative method gathering data points in clusters using a distance metric representing their dissimilarity.” reads on clustering distance. )
It would be obvious to one of the ordinary skill in the art before the effective filing date of claim invention to combine both Euclidean and Clustering distance as the combination benefits from topological mapping (SOM), geometric precision (Euclidean), and domain‑specific similarity (clustering distance), resulting in clusters that are both structurally sound and interpretably meaningful.
The combination of Hazrati, Yao and Tamas, does not teach “performing weighted calculation on the Euclidean distance and the clustering distance to determine a winning neuron;”
Wehrens teaches “performing weighted calculation on the Euclidean distance and the clustering distance to determine a winning neuron”. (Wehrens, page 3
PNG
media_image1.png
63
434
media_image1.png
Greyscale
reads on weighted calculation of Euclidean and Clustering distance to determine winning neuron)
Hazrati teaching on Euclidean distance and Tamas teaching of clustering distance in view of Wehrens’s weighted combined distance calculation results improve accuracy in pollution source detection.
Therefore, it would be obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Hazrati and Tamas with Wehrens weighted sum to Euclidean and clustering distance in benefits from integrating geometric and structural information, leading to more accurate, stable, and interpretable clustering results and provides robustness to outliers and scale difference.
Hazrati further teaches “updating the output neuron weight vector of the water pollution neural network according to an input neuron weight vector and the output neuron weight vector of the winning neuron; and when a quantity of updating times of the output neuron weight vector of the water pollution neural network reaches a preset value, determining the groundwater pollution source of the target area by the updated water pollution neural network”(Hazrati, page 8 “once the winner neuron is obtained”, “the weight vector of the winning neuron and all other neurons are updated according to Equation (4) to minimize the local error”
PNG
media_image2.png
38
481
media_image2.png
Greyscale
reads on “updating the output neuron weight vector of the water pollution neural network according to an input neuron weight vector and the output neuron weight vector of the winning neuron”;
“Adaptation: The weight adjusting is repeated until a stable map is obtained
or the map is converged” Pg 4, para 5, reads on “when a quantity of updating times of the output neuron weight vector of the water pollution neural network reaches a preset value, determining the groundwater pollution source of the target area by the updated water pollution neural network.”)
Hazrati teaches groundwater contaminant source characterization with SOM, Yao teaches surface water system and enterprise type data collection for pollution source , Tamas teaches calculating Euclidean and clustering distance from sample data and Wherens teaches adding weights to multiple distance.
Therefore, it would have been obvious for someone of ordinary skilled in the art to combine the teaching of Hazrati, Yao, Tamas, Wehrens enabling output neuron weight vector of the water pollution neural network to be jointly updated according to the Euclidean distance and the clustering distance. This result from integrating geometric and structural information, leading to more accurate, stable, and interpretable clustering results and provides robustness to outliers and scale difference in neural network and pollution source detection.
Claim 2, Hazrati further teaches “wherein calculating the Euclidean distance between the sample data and the corresponding output neuron weight vector includes: calculating the Euclidean distance between the sample data and the output neuron using the following formula: dxj=∑i=1nwji-xji2 where dxj is the Euclidean distance between the jth sample data and the output neuron, j∈[1,N], N is the number of the sample data, wji is the output neuron weight vector, xji is the input neuron weight vector corresponding to the ith data dimension in the jth sample data, n is the data dimension of the sample data.” (page 8 , 1std paragraph
PNG
media_image3.png
190
696
media_image3.png
Greyscale
reads on formula: dxj=∑i=1nwji-xji2).
Therefore, it would have been obvious for one skilled in the ordinary art before the effective filing date of the claimed invention to add the teaching of Euclidean distance calculation of Hazrati in view of Yao, Tamas, and Wehrens to improve accuracy of the groundwater pollution detection neural network.
Claims 3-4 are rejected under 35 U.S.C. 103 as being unpatentable over Hazrati, Yao, Tamas, Wehrens and further in view of Sudipto et al, CURE: An efficient Clustering Algorithm for Large Database, hereinafter Sudipto.
Claim 3, The combination of Hazrati, Tamas and Wehrens does not teach “wherein calculating the clustering distance between the sample data and the corresponding output neuron weight vector includes: calculating the clustering distance between the sample data and the output neuron using the following formula: distxj,wj=xj-wj where distxj,wj is the clustering distance between the sample data and the output neuron, j∈[1,N],N is the number of the sample data, xj is the jth sample data, and wji is the jth output neuron.”
Sudipto teaches “calculating the clustering distance between the sample data and the corresponding output neuron weight vector includes: calculating the clustering distance between the sample data and the output neuron using the following formula: distxj,wj=xj-wj where distxj,wj is the clustering distance between the sample data and the output neuron, j∈[1,N],N is the number of the sample data, xj is the jth sample data, and wji is the jth output neuron.”( Sudipto, page 5 section 3.2 reads on
PNG
media_image4.png
269
516
media_image4.png
Greyscale
calculating the clustering distance between the sample data and the output neuron.
Note: CURE hierarchical Clustering algorithm taught by Sudipto can discover clusters with interesting shapes and less sensitive to outliers with low execution time (Sudipto, page 8 Col 2, 1st para, which is the same approach of this application in light of the specification [0022]).
Therefore, it would have been obvious for one skilled in the ordinary art before the effective filing date of the claimed invention to combine the teaching of Hazrati with Sudipto’s CURE clustering distance in view of Yao, Tamas, Wehrens to improve sensitivity to outlier and low execution time of the groundwater pollution neural network.
Claim 4, Wehrens teaches “wherein performing weighted calculation on the Euclidean distance and the clustering distance to determine the winning neuron includes”;( Wehrens, page 3, discussed in claim 1)
The combination of Hazrati, Tamas and Wehrens does not teach “obtaining a minimum Euclidean distance and a minimum clustering distance; and performing weighted calculation on the minimum Euclidean distance and the minimum clustering distance to determine the winning neuron.”
Sudipto teaches “obtaining a minimum Euclidean distance and a minimum clustering distance;”(pg 5, section 3.2
PNG
media_image5.png
73
467
media_image5.png
Greyscale
PNG
media_image6.png
40
386
media_image6.png
Greyscale
reads on minimum Euclidean and minimum clustering distance.
Hazrati’s teaching of Euclidean distance in SOM combined with Wehrens’s teaching of using weighted contribution of multiple distance and Sudipto’s teaching of minimum distance results weighted minimum Euclidean and minimum clustering distance.
Therefore, it would have been obvious for one skilled in the ordinary art before the effective filing date of the claimed invention to modify Hazrati with Wehrens weighted sum with Sudipto’s minimum Euclidean and minimum clustering distance yielding predictable result. (ref KSR Int’l Co. v. Teleflex Inc.)
Claims 5,6,7 are rejected as under 35 U.S.C. 103 as being unpatentable over Hazrati, in view of Yao, Tamas, Wehrens, Sudipto and further in view of NPL ref: Some measures to impact on the performance of Kohonen self-organizing map, Vijaya et al hereinafter Vijaya.
Claim 5 teaches, “wherein updating the output neuron weight vector of the water pollution neural network according to the input neuron weight vector and the output neuron weight vector of the winning neuron includes”; (Hazrati, page 8 as discussed in claim 1).
Hazrati, Yao, Wehrens, Tamas, Sudipto do not teach, “obtaining the influence range of the winning neuron; and updating the output neuron weight vector of the water pollution neural network according to the influence range, the input neuron weight vector, and the output neuron weight vector of the winning neuron”.
Vijaya teaches “wherein obtaining the influence range of the winning neuron; and updating the output neuron weight vector of the water pollution neural network according to the influence range, the input neuron weight vector, and the output neuron weight vector of the winning neuron”. (Vijaya, page 4
PNG
media_image7.png
434
417
media_image7.png
Greyscale
section 2.1).
So, it would have been obvious for one of ordinary skill in the art before the effective filing date of the claimed invention to combine Hazrati’s teaching with Vijaya’s influence range to improve robustness, accuracy and quality of pollution detection neural network.
Claim 6, Hazrati, Yao, Sudipto, Tamas and Wehrens do not teach “wherein obtaining the influence range of the winning neuron includes: calculating the influence range of the winning neuron using the following formula: σt=σ0ⅇ-tτ0 where σt is the influence range of the winning neuron, σ0 is the initial influence range of the winning neuron, t is the time length, τ0 is the fixed attenuation coefficient.”
Vijaya teaches “wherein obtaining the influence range of the winning neuron includes: calculating the influence range of the winning neuron using the following formula: σt=σ0ⅇ-tτ0 where σt is the influence range of the winning neuron, σ0 is the initial influence range of the winning neuron, t is the time length, τ0 is the fixed attenuation coefficient.”(Vijaya, page 4
PNG
media_image8.png
167
560
media_image8.png
Greyscale
NOTE: Conventional SOM teaches exponential decaying neighborhood radius or influence range. )
SOM neighborhood update (influence range) gives the map topology preservation, robustness, error and quality improvement.
It would been obvious for someone of ordinary skill in the art before the effective filing date of the claimed invention to add Vijaya’s influence range for error and quality improvement.
Claim 7, Hazrati , Yao, Tamas, Wehrens and Sudipto do not teach “wherein updating the output neuron weight vector of the water pollution neural network using the following formula: Δwp=ηt*Tt*xp-wp ηt=η0ⅇ-tτη Tt=ⅇ-dpq22σt2 where Δwp is the updated output neuron weight vector of the water pollution neural network, ηt is the learning coefficient that is attenuated over time, Tt is the neighborhood influence coefficient, η0 is the initial value of the learning coefficient, xp is the input neuron weight vector of the winning neuron, wp is the output neuron weight vector of the winning neuron, dpq is the Euclidean distance between the winning neuron and its neighbor neuron, τη is the fixed learning coefficient, σt is the influence range of the winning neuron.”
Vijaya further teaches “wherein updating the output neuron weight vector of the water pollution neural network using the following formula: Δwp=ηt*Tt*xp-wp ηt=η0ⅇ-tτη Tt=ⅇ-dpq22σt2 where Δwp is the updated output neuron weight vector of the water pollution neural network, ηt is the learning coefficient that is attenuated over time, Tt is the neighborhood influence coefficient, η0 is the initial value of the learning coefficient, xp is the input neuron weight vector of the winning neuron, wp is the output neuron weight vector of the winning neuron, dpq is the Euclidean distance between the winning neuron and its neighbor neuron, τη is the fixed learning coefficient, σt is the influence range of the winning neuron.” (Vijaya, page 4-5
PNG
media_image9.png
307
913
media_image9.png
Greyscale
PNG
media_image10.png
199
877
media_image10.png
Greyscale
PNG
media_image11.png
398
698
media_image11.png
Greyscale
Eq 5 reads on Tt=ⅇ-dpq22σt2 , Eq 4 reads on Δwp=ηt*Tt*xp-wp
Formula in claim 7 is known in Vijaya’s Kohonen self-organizing map (SOM) ).
Hazrati teaches groundwater contamination source characterization using Self Organizing Map but does not teach time dependent learning rate.
It would be obvious for someone of ordinary skill in the art before the effective filing date of the claimed invention to apply known technique of Vijaya by controlling magnitude and neighborhood extent of SOM weight adaptation during training. Applying this SOM technique to Hazrati’s SOM would have provided time dependent learning and neighborhood adjustment with existing SOM weight update operation.
Conclusion
All claims are rejected on under 35 U.S.C. 103 . The prior art made of record and not relied upon is considered pertinent to applicant's disclosure, Archives of Environmental Contamination and Toxicology (2021) 81:397–413, Groundwater Pollution Source Identification and Apportionment Using PMF and PCA‑APCS‑MLR Receptor Models in Tongchuan City, China. PCA-APCS-MLR and the limitation of linear method limitation was considered.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MOIN UDDIN AHMED whose telephone number is (571)270-3785. The examiner can normally be reached 8am-5pm ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, EMAN ALKAFAWI can be reached 571-272-4448. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Mon Ahmed/
Paten Examiner
Art Unit 2858
09/11/2026
/EMAN A ALKAFAWI/Supervisory Patent Examiner, Art Unit 2858 9/17/2026
Prosecution Insights