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 .
This action is in response to Application filed 08/21/2025.
Claims 1-20 are pending.
Priority
This application claims priority from U.S. Provisional Application No. 63/693,495 filed on 09/11/2024. The provisional application provides sufficient support as requirements under 35 U.S.C. § 112(a) or (pre-AIA ) 35 U.S.C. § 112, first paragraph for claims 1-2, 4, 6-13, 15 and 17-20, but not for claims 3, 5, 14 and 16. Therefore, claims 1-2, 4, 6-13, 15 and 17-20 have an effective filing data of 09/11/2024, and claims 3, 5, 14 and 16 have an effective filing date of 08/21/2025.
Information Disclosure Statement
The Information Disclosure Statement (IDS) filed by Applicant on 08/25/2025 has been considered. A copy of the considered IDS is enclosed with this Office action.
Specification
The disclosure is objected to because of the following informalities:
In view of the replacement sheets of drawings filed on 09/12/2025 including Fig. 5A and Fig. 5B, there are no references for Fig. 5A and Fig. 5B in the Specification. It is suggested that Applicant can specify Fig. 5 in the “Brief Description of the Drawings” section as including Fig. 5A and Fig. 5B in order to overcome this objection.
Appropriate correction is required.
The specification is objected to as failing to provide proper antecedent basis for the claimed subject matter. See 37 CFR 1.75(d)(1) and MPEP § 608.01(o). Correction of the following is required: the limitation “a memory component” in line 3 of claim 15 and the limitation “a processor component” in line 5 and line 14 of claim 15 and in line 1 of claim 20.
Claim Objections
Claim 2 is objected to because of the following informalities:
Regarding claim 2, the term “the vector space” in line 2 should be “the high-dimensional vector space” for being consistent in claim language.
Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation "the selected top data points" in line 19. There is insufficient antecedent basis for this limitation in the claim.
Claim 15 recites the limitation “the relevance calculator” in line 24 and the limitation "the selected top data points" in line 24 There are insufficient antecedent basis for this limitation in the claim.
In addition, claims 1 and 15 recite a search process based on a user query. However, it is unclear regarding how a search result is generated to response to the user query. The recitation of performing a sequential cluster search as recited does not realize the benefit of the invention (e.g., whether all clusters are being searched or only a subset of clusters is being searched). As claimed, each cluster is being searched and there is no limit on a number of clusters being identified/selected for searching based on Euclidean distances between the prompt embedding (i.e., query vector) and a plurality of cluster centers (i.e., representing clusters of data points in the high-dimensional vector space). In addition, it is unclear how each set of top data points from searching a cluster and semantic relevance score are used in filtering and/or combining with other sets of top data points from other clusters in order to generate a search result for the user query. Therefore, the metes and bounds of the claimed invention is unclear.
Regarding claims 8 and 18, it is unclear how “the semantic relevance scores” can be computed based on the calculation of a first boundary point and a second boundary point as recited (e.g., how the first boundary point and/or the second boundary point are related to a particular distance and a particular semantic relevance score). Also, symbol α and symbol β are not specified.
Other dependent claims are rejected as incorporating and failing to resolve the deficiencies of rejected independent claims 1 and 15 upon which they depend correspondingly.
Claim limitations “a distance calculation module”, “a cluster module” and “a relevance scoring module” in claim 15 invokes 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. However, the written description fails to disclose the corresponding structure, material, or acts for performing the entire claimed function and to clearly link the structure, material, or acts to the function. It is unclear how each module as stated above is implemented in combination of particular structure (i.e., hardware) and algorithm (i.e., software) in order to perform the function as recited by each module. Therefore, the claim is indefinite and is rejected under 35 U.S.C. 112(b) or pre-AIA 35 U.S.C. 112, second paragraph.
Applicant may:
(a) Amend the claim so that the claim limitation will no longer be interpreted as a limitation under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph;
(b) Amend the written description of the specification such that it expressly recites what structure, material, or acts perform the entire claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(c) Amend the written description of the specification such that it clearly links the structure, material, or acts disclosed therein to the function recited in the claim, without introducing any new matter (35 U.S.C. 132(a)).
If applicant is of the opinion that the written description of the specification already implicitly or inherently discloses the corresponding structure, material, or acts and clearly links them to the function so that one of ordinary skill in the art would recognize what structure, material, or acts perform the claimed function, applicant should clarify the record by either:
(a) Amending the written description of the specification such that it expressly recites the corresponding structure, material, or acts for performing the claimed function and clearly links or associates the structure, material, or acts to the claimed function, without introducing any new matter (35 U.S.C. 132(a)); or
(b) Stating on the record what the corresponding structure, material, or acts, which are implicitly or inherently set forth in the written description of the specification, perform the claimed function. For more information, see 37 CFR 1.75(d) and MPEP §§ 608.01(o) and 2181.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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, 4-7, 10-11, 13-15 and 17 (effective filing date 09/11/2024) are rejected under 35 U.S.C. 103 as being unpatentable over Bortnikov et al. (U.S. Publication No. 2017/0140012, Publication date 05/18/2017), in view of Li et al. (U.S. Publication No. 2020/0250538, Publication date 08/06/2020), and further in view of Zhao et al. (CN-110874385-B, Publication date 11/14/2023).
As to claim 1, Bortnikov et al. teaches:
“A computer-implemented method for semantic search optimization in high-dimensional vector spaces” (see Bortnikov et al., Abstract), the method comprising:
“(A) receiving a user query” (see Bortnikov et al., [0043] for receiving a query);
“(B) generating a prompt embedding from the user query, wherein the prompt embedding comprises a numerical vector in a high-dimensional vector space” (see Bortnikov et al., [0043]-[0044] for generating a vector of numerical values or a high dimensional vector to represent the query, wherein the vector of numerical values or a high dimensional vector as disclosed is interpreted as a prompt embedding as recited);
“(C) determining Euclidean distances between the prompt embedding and a plurality of cluster centers in the high-dimensional vector space, wherein each cluster center represents a cluster of data points” (see Bortnikov et al., [0044] for determining the distance between the query and each of the cluster centroids in the original space (i.e., vector space), wherein each of data points (e.g., the query or the cluster centroids) are represented as a d-dimensional vectors or high-dimensional vectors and the distance metric is the Euclidean distance);
“(D) identifying a jump point as the cluster center having the shortest Euclidean distance to the prompt embedding” (see Bortnikov et al., [0045] for identifying the subset of the clusters that are closest to the query based on determining the distance between the query and each of the cluster centroids; it should be noted that based on determining the distance between the query and each of the cluster centroids, the cluster centroid (i.e., a jump point) having the shortest/closest distance to the query must be identified);
“(E) performing a sequential cluster search starting from the jump point and proceeding in ascending order of Euclidean distance from the prompt embedding, the sequential cluster search comprising, for each cluster searched” (see Bortnikov et al., [0047]-[0048] for processing/searching each cluster of the identified subset of clusters):
“(E)(1) performing a K-Nearest Neighbor (KNN) search within the cluster to determine Euclidean distances between the prompt embedding and data points within the cluster” (see Bortnikov et al., [0040] and [0047]-[0048] for determining distances between the query and each data point in each cluster ; also see [0002] for using Euclidean distance);
“(E)(2) selecting a set of top data points having the shortest Euclidean distances to the prompt embedding” (see Bortnikov et al., [0050] for identifying a set of nearest neighbors (i.e., data points within the cluster that are closest to the query) using the distance table for a corresponding cluster).
However, Bortnikov et al. does not explicitly teach a feature of computing relevance scores based on Euclidean distances as equivalently recited as follows:
“(E)(3) computing semantic relevance scores for the selected top data points using a relevance function based on the determined Euclidean distances”.
On the other hand, Li et al. explicitly teaches a feature of computing relevance scores for data points/items based on Euclidean distances (see Li et al., [0087] for determining a respective relevance score for a given image based on a measure of similarity (e.g., a Euclidean distance) between the embedding of the given image and the embedding of the query).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to incorporate Li et al.'s teaching to Bortnikov et al.’s system by implementing a feature of determining relevance scores for data points/items based on Euclidean distances. Skilled artisan would have been motivated to do so, as suggested by Li et al. (see [0087]), to provide Bortnikov et al.’s system with an effective way for ranking search results for a search query. In addition, both of the references (Bortnikov et al. and Li et al.) teach features that are directed to analogous art and they are directed to the same field of endeavor, such as, a search system based on embeddings/vectors. This close relation between both of the references highly suggests an expectation of success when combined.
In case that Bortnikov et al. as modified by Li et al. does not explicitly teach a feature of sequentially searching/processing clusters based on the distances between the query and theirs corresponding cluster centroids in an ascending order as recited as follows:
“(E) performing a sequential cluster search starting from the jump point and proceeding in ascending order of Euclidean distance from the prompt embedding, the sequential cluster search comprising, for each cluster searched.”
Zhao et al. explicitly teaches a feature of sequentially searching/processing clusters based on the distances between the query and theirs corresponding cluster centroids in an ascending order (see Zhao et al., [page 6, lines 30-39] and [page 13, lines 6-10] the first Hamming distance between the query vector and each cluster central point can be calculated, and according to the order from small to large (i.e., distance), screening/processing/searching at least one first vector set/cluster).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to incorporate Zhao et al.'s teaching to Bortnikov et al.’s system (as modified by Li et al.) by implementing a feature of sequential searching/processing of clusters according to ascending order (i.e., the order from small to large). Skilled artisan would have been motivated to do so to provide Bortnikov et al.’s system with an effective and alternative way for searching clusters. In addition, both of the references (Bortnikov et al. and Zhao et al.) teach features that are directed to analogous art and they are directed to the same field of endeavor, such as, a search system based on embeddings/vectors and clustering embeddings/vectors. This close relation between both of the references highly suggests an expectation of success when combined.
As to claim 4, this claim is rejected based on the same arguments as above to reject claim 4 and is similarly rejected including the following:
Bortnikov et al. as modified by Li et al. and Zhao et al. teaches:
“calculating the Euclidean distances between the prompt embedding and each cluster center utilizes a first distance metric” (see Bortnikov et al., [0045] for determining/calculating the distance between the query and each of the cluster centroids; also see [0002] for the distance metric is the Euclidean); and
“computing the semantic relevance scores employs a second, different, distance metric” (see Bortnikov et al., [0049] for other distance metrics; also see Li et al., [0087] and [0070] for determining the relevance score based on a measure of similarity (e.g., cosine similarity measure)).
As to claim 5, this claim is rejected based on the same arguments as above to reject claim 1 and is similarly rejected including the following:
Bortnikov et al. as modified by Li et al. and Zhao et al. teaches:
“wherein performing the search for data points is performed within a maximum of 10 milliseconds for 1 million samples” (see Bortnikov et al., [0003] and [0072] for improved speed using GPU).
As to claim 6, this claim is rejected based on the same arguments as above to reject claim 5 and is similarly rejected including the following:
Bortnikov et al. as modified by Li et al. and Zhao et al. teaches:
“wherein the search for data points processes at least 100,000 data points” (see Bortnikov et al., [0020] and [0072] for performing search on large data sets).
As to claim 7, this claim is rejected based on the same arguments as above to reject claim 1 and is similarly rejected including the following:
Bortnikov et al. as modified by Li et al. and Zhao et al. teaches:
“wherein using the KNN search algorithm comprises using an optimized similarity search library for enhanced performance” (see Bortnikov et al., [0037] for partitioning/clustering a dataset into a plurality of clusters and storing the results of partitioning/clustering (e.g., clusters, cluster centroids) for retrieval during a subsequent kNN search, wherein a storage of clustering data as disclosed can be interpreted as an optimized similarity search library as recited).
As to claim 10, this claim is rejected based on the same arguments as above to reject claim 1 and is similarly rejected including the following:
Bortnikov et al. as modified by Li et al. and Zhao et al. teaches:
“wherein the high-dimensional vector space comprises at least one hundred thousand embeddings” (see Bortnikov et al., [0019]-[0020] and [0072] for performing search on large data sets including data points wherein each data point represented by a vector/embedding).
As to claim 11, this claim is rejected based on the same arguments as above to reject claim 1 and is similarly rejected including the following:
Bortnikov et al. as modified by Li et al. and Zhao et al. teaches:
“wherein the prompt embedding has at least 768 dimensions” (see Bortnikov et al., [0044] for a high dimensional vector (i.e., prompt embedding) representing a query).
As to claim 13, this claim is rejected based on the same arguments as above to reject claim 1 and is similarly rejected including the following:
Bortnikov et al. as modified by Li et al. and Zhao et al. teaches:
“further comprising inserting the selected set of top data points having the shortest Euclidean distances to the prompt embedding as relevant context from the high-dimensional vector space into a prompt for a large language model, and wherein the relevant context is based on the computed semantic relevance scores” (see Bortnikov et al., [0018] for retrieval of search results, which can be used for any purpose, wherein the query represents a relevant context).
As to claim 14, this claim is rejected based on the same arguments as above to reject claim 1 and is similarly rejected including the following:
Bortnikov et al. as modified by Li et al. and Zhao et al. teaches:
“calculating Euclidean distances between the prompt embedding and the plurality of cluster centers comprises implementing distributed cluster processing by simultaneously calculating the Euclidean distances for different subsets of cluster centers across multiple computational nodes”(see Bortnikov et al., [0045] for ascertain/determining the distance between the query and each of the cluster centroids/centers, wherein the distance computations are performed in parallel by multiple CPUs (i.e., multiple computational nodes)); and
“performing the sequential cluster search comprises processing different clusters simultaneously across the multiple computational nodes” (see Bortnikov et al., [0045] for performing distance calculations in parallel by multiple CPUs (i.e., the multiple computational nodes)).
As to claim 15, Bortnikov et al. teaches:
“A system for semantic search optimization in high-dimensional vector spaces” (see Bortnikov et al., Abstract), the system comprising:
“a memory component configured to store a database of data points organized into clusters in a high-dimensional vector space” (see Bortnikov et al., [0037] for partitioning/clustering a data set (i.e., data points) into a plurality of clusters and storing the results of the partitioning/clustering, wherein any storage for the data set and/or results of the partitioning/clustering can be interpreted as a memory component as recited);
“a processor component configured to generate a prompt embedding from a user query, wherein the prompt embedding comprises a numerical vector in the high-dimensional vector space” (see Bortnikov et al., [0043]-[0044] for generating a vector of numerical values or a high dimensional vector to represent the query, wherein the vector of numerical values or a high dimensional vector as disclosed is interpreted as a prompt embedding as recited, and any code/module for generating the query vector as disclosed can be interpreted as equivalent to a processor component as recited);
“a distance calculation module configured to determine Euclidean distances between vectors in the high-dimensional vector space” (see Bortnikov et al., [0002], [0045] and [0047] for determining the distance (e.g., Euclidean distance) between a query and each of cluster centroids and/or between a query and each of data points, wherein any code/module for determining/calculating the distance as disclosed can be interpreted as a distance calculation module as recited);
“a clustering module configured to organize the data points into clusters, wherein each cluster center represents a cluster of data points” (see Bortnikov et al., [0037] for partitioning/clustering a data set (i.e., data points) into a plurality of clusters, wherein any code/module for partitioning/clustering as disclosed can be interpreted as equivalent to a clustering module as recited);
“a search engine configured to perform K-Nearest Neighbor (KNN) searches within clusters” (see Bortnikov et al., [0040] and [0050] wherein the hardware accelerator as disclosed can be interpreted as equivalent to the search engine as recited); and
“wherein the processor component is further configured to” (see Bortnikov et al., [0026] and [0043] for a processor of a computing device):
“identify a jump point as the cluster center having the shortest Euclidean distance to the prompt embedding” (see Bortnikov et al., [0045] for identifying the subset of the clusters that are closest to the query based on determining the distance between the query and each of the cluster centroids; it should be noted that based on determining the distance between the query and each of the cluster centroids, the cluster centroid (i.e., a jump point) having the shortest/closest distance to the query must be identified);
“perform a sequential cluster search starting from the jump point and proceeding in ascending order of Euclidean distance from the prompt embedding” (see Bortnikov et al., [0047]-[0048] for processing/searching each cluster of the identified subset of clusters):
“for each cluster searched, utilize the search engine to perform a KNN search within the cluster to determine Euclidean distances between the prompt embedding and data points within the cluster” (see Bortnikov et al., [0040] and [0047]-[0048] for determining distances between the query and each data point in each cluster ; also see [0002] for using Euclidean distance);
“select a set of top data points having the shortest Euclidean distances to the prompt embedding” (see Bortnikov et al., [0050] for identifying a set of nearest neighbors (i.e., data points within the cluster that are closest to the query) using the distance table for a corresponding cluster).
However, Bortnikov et al. does not explicitly teach a feature of computing relevance scores based on Euclidean distances as equivalently recited as follows:
“a relevance scoring module configured to compute semantic relevance scores using a relevance function based on the determined Euclidean distances; and
utilize the relevance calculator to compute semantic relevance scores for the selected top data points”.
On the other hand, Li et al. explicitly teaches a feature of computing relevance scores for data points/items based on Euclidean distances (see Li et al., [0087] for determining a respective relevance score for a given image based on a measure of similarity (e.g., a Euclidean distance) between the embedding of the given image and the embedding of the query).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to incorporate Li et al.'s teaching to Bortnikov et al.’s system by implementing a feature of determining relevance scores for data points/items based on Euclidean distances. Skilled artisan would have been motivated to do so, as suggested by Li et al. (see [0087]), to provide Bortnikov et al.’s system with an effective way for ranking search results for a search query. In addition, both of the references (Bortnikov et al. and Li et al.) teach features that are directed to analogous art and they are directed to the same field of endeavor, such as, a search system based on embeddings/vectors. This close relation between both of the references highly suggests an expectation of success when combined.
In case that Bortnikov et al. as modified by Li et al. does not explicitly teach a feature of sequentially searching/processing clusters based on the distances between the query and theirs corresponding cluster centroids in an ascending order as recited as follows:
“perform a sequential cluster search starting from the jump point and proceeding in ascending order of Euclidean distance from the prompt embedding.”
Zhao et al. explicitly teaches a feature of sequentially searching/processing clusters based on the distances between the query and theirs corresponding cluster centroids in an ascending order (see Zhao et al., [page 6, lines 30-39] and [page 13, lines 6-10] the first Hamming distance between the query vector and each cluster central point can be calculated, and according to the order from small to large (i.e., distance), screening/processing/searching at least one first vector set/cluster).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to incorporate Zhao et al.'s teaching to Bortnikov et al.’s system (as modified by Li et al.) by implementing a feature of sequential searching/processing of clusters according to ascending order (i.e., the order from small to large). Skilled artisan would have been motivated to do so to provide Bortnikov et al.’s system with an effective and alternative way for searching clusters. In addition, both of the references (Bortnikov et al. and Zhao et al.) teach features that are directed to analogous art and they are directed to the same field of endeavor, such as, a search system based on embeddings/vectors and clustering embeddings/vectors. This close relation between both of the references highly suggests an expectation of success when combined.
As to claim 17, this claim is rejected based on the same arguments as above to reject claim 15 and is similarly rejected including the following:
Bortnikov et al. as modified by Li et al. and Zhao et al. teaches:
“calculating the Euclidean distances between the prompt embedding and each cluster center utilizes a first distance metric” (see Bortnikov et al., [0045] for determining/calculating the distance between the query and each of the cluster centroids; also see [0002] for the distance metric is the Euclidean); and
“computing the semantic relevance scores employs a second, different, distance metric” (see Bortnikov et al., [0049] for other distance metrics; also see Li et al., [0087] and [0070] for determining the relevance score based on a measure of similarity (e.g., cosine similarity measure)).
Claim 2 (effective filing date 09/11/2024) is rejected under 35 U.S.C. 103 as being unpatentable over Bortnikov et al. (U.S. Publication No. 2017/0140012, Publication date 05/18/2017), in view of Li et al. (U.S. Publication No. 2020/0250538, Publication date 08/06/2020), in view of Zhao et al. (CN-110874385-B, Publication date 11/14/2023), and further in view of Dehaspe et al. (US Publication No. 2020/0080158, Publication date 03/12/2020).
As to claim 2, Bortnikov et al. as modified by Li et al. and Zhao et al. teaches all limitations as recited in claim 1.
However, Bortnikov et al. as modified by Li et al. and Zhao et al. does not explicitly teach a feature of determining a number of clusters based on a square root of a total number of data points as recited as follows:
“wherein (C) comprises determining a number of clusters in the high-dimensional vector space based on a square root of a total number of data points in the vector space”.
On the other hand, Dehaspe et al. explicitly teaches a feature of determining a number of clusters based on a square root of a total number of data points (see Dehaspe et al., [0242] for determining a number of clusters that approximates the square root of the total number of samples (i.e., data points)).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to incorporate Dehaspe et al.'s teaching to Bortnikov et al.’s system (as modified by Li et al. and Zhao et al.) by implementing a feature of determining a number of clusters for clustering based on the square root of the total number of data points. Skilled artisan would have been motivated to do so to provide Bortnikov et al.’s system with an effective way for clustering data points. In addition, both of the references (Bortnikov et al. and Dehaspe et al.) teach features that are directed to analogous art and they are directed to the same field of endeavor, such as, clustering data points into clusters. This close relation between both of the references highly suggests an expectation of success when combined.
Claims 3 and 16 (effective filing date 09/11/2024) is rejected under 35 U.S.C. 103 as being unpatentable over Bortnikov et al. (U.S. Publication No. 2017/0140012, Publication date 05/18/2017), in view of Li et al. (U.S. Publication No. 2020/0250538, Publication date 08/06/2020), in view of Zhao et al. (CN-110874385-B, Publication date 11/14/2023), and further in view of Sawarkar et al. (US Publication No. 2022/0138786, Publication date 05/05/2022).
As to claims 3 and 16, Bortnikov et al. as modified by Li et al. and Zhao et al. teaches all limitations as recited in claims 1 and 15 respectively.
However, Bortnikov et al. as modified by Li et al. and Zhao et al. does not explicitly teach a feature of determining a number of clusters based on density, distribution and other performance metrics as recited as follows:
“determining a number of clusters in the high- dimensional vector space, and wherein determining the number of clusters in the high-dimensional vector space comprises:
analyzing vector space density characteristics;
evaluating embedding distribution patterns; and
dynamically adjusting the number of clusters based on real-time performance metrics”.
On the other hand, Sawarkar et al. explicitly teaches a feature of determining a number of clusters based on density, distribution pattern and other performance metrics (see Sawarkar et al., [0070] for determining a number of clusters based on value of epsilon (i.e., density or distribution pattern) and other performance metrics (e.g., revenue, profit, efficiency, etc.)).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to incorporate Sawarkar et al.'s teaching to Bortnikov et al.’s system (as modified by Li et al. and Zhao et al.) by implementing a feature of determining a number of clusters based on density, distribution pattern and other performance metrics. Skilled artisan would have been motivated to do so to provide Bortnikov et al.’s system with an effective way for clustering data points. In addition, both of the references (Bortnikov et al. and Sawarkar et al.) teach features that are directed to analogous art and they are directed to the same field of endeavor, such as, clustering data points into clusters. This close relation between both of the references highly suggests an expectation of success when combined.
Claims 12 and 20 (effective filing date 09/11/2024) is rejected under 35 U.S.C. 103 as being unpatentable over Bortnikov et al. (U.S. Publication No. 2017/0140012, Publication date 05/18/2017), in view of Li et al. (U.S. Publication No. 2020/0250538, Publication date 08/06/2020), in view of Zhao et al. (CN-110874385-B, Publication date 11/14/2023), and further in view of Yang et al. (US Publication No. 2004/0017947, Publication date 01/29/2004).
As to claims 12 and 20, Bortnikov et al. as modified by Li et al. and Zhao et al. teaches all limitations as recited in claims 1 and 15 respectively including identifying a set of nearest neighbors (i.e., a set of relevant data points) (see Bortnikov et al., [0050]; also see Li et al., [0087]).
In addition, Bortnikov et al. as modified by Li et al. and Zhao et al. further teaches:
“fetch the identified relevant data points” (see Bortnikov et al., [0018] and [0051] for retrieval of search results (e.g., similar images) in response to a query).
However, Bortnikov et al. as modified by Li et al. and Zhao et al. does not explicitly teach a feature of identifying relevant data points/items based on a specified distance threshold as recited as follows:
“identify relevant data points within a specified distance threshold based on the computed semantic relevance scores”.
On the other hand, Yang explicitly teaches a feature of identifying relevant data points/items based on a specified distance threshold (see Yang, claim 4 for identifying/selecting data points within a predetermined radius (i.e., a specified distance threshold) based on the determined distance).
It would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention to incorporate Yang's teaching to Bortnikov et al.’s system (as modified by Li et al. and Zhao et al.) by implementing a feature of identifying relevant data points/items within a distance threshold. Skilled artisan would have been motivated to do so to provide Bortnikov et al.’s system with an effective way for selecting relevant data points. In addition, using a threshold based on determined score/value is well-known and well-used in the art for selecting and/or limiting the selected data items.
Allowable Subject Matter
Claims 8-9 and 18-19 would be allowable if rewritten to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph, set forth in this Office action and to include all of the limitations of the base claim and any intervening claims.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PHUONG THAO CAO whose telephone number is (571)272-2735. The examiner can normally be reached Monday - Friday: 9:00 am - 6:00 pm.
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, Amy Ng can be reached at 571-270-1698. 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.
/Phuong Thao Cao/Primary Examiner, Art Unit 2164
Prosecution Insights