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 .
Status of Claims
The office action is being examined in response to the application filed by the applicant on April 28th, 2026.
Claims 1, 8, 15, and 18 – 20 have been amended and are hereby entered.
Claims 2, 3, 8, 10, 16 and 17 have been cancelled.
Claims 1, 4 – 8, 11 – 15, and 18 - 20 are pending and have been examined.
This action is made FINAL.
Response to Arguments
Applicant’s arguments filed on April 28th, 2026 have been fully considered but are not persuasive.
Regarding Applicant’s arguments against the 101 rejections on p. 10 – 21: Applicant argues that amended claim 1 no longer recites an abstract idea because the incorporated data-cleansing and feature-extraction limitations cannot be practically performed in the human mind. The Examiner respectfully disagrees. While certain newly added calculations need not be characterized as mental processes, they expressly recite mathematical concepts, including calculating neighborhood distances and mean values, calculating matrix elements, according to a mathematical formula, generating a probability transition matrix, and calculating similarity according to a recited mathematical formula. The remaining limitations of defining relationships, evaluating user characteristics according to personal interests, friend relationships, and community influences, and predicting user relationships constitute evaluations, judgments, and conclusions concerning information and human relationships. Thus, considered together, the claim continues to recite an abstract idea comprising mathematical analysis and evaluation of user information to predict user relationships.
Applicant’s arguments regarding Step 2A, Prong Two are likewise unpersuasive. Although Applicant characterizes the amendments as a technological data-cleansing pipeline that improves prediction accuracy, the claimed improvement is to the accuracy of the resulting user-relationship prediction, rather than to the functioning of a computer or another technology. Converting data into a database, generating a database object set, performing file parsing, and processing abnormal data merely prepare and organize information for the claimed mathematical analysis. Further, claim 1 does not recite a processor, computer, specialized database architecture, or other technological component that applies the abstract idea in a manner that provides a technological improvement. Applicant’s reliance on unclaimed techniques such as topic modeling, interest graphs, interaction-frequency measurements, and community-detection algorithms is not persuasive because those implementations are not required by the claim. Accordingly, the judicial exception is not integrated into a practical application.
Applicant’s arguments regarding Step 2B are also unpersuasive. The neighborhood-distance calculations, probability transition matrix, similarity calculations, feature analysis, relationship modeling, and prediction relied upon by Applicant constitute the identified abstract idea and cannot themselves supply the additional elements amounting to significantly more. The remaining database conversion, database-object generation, and file-parsing operations merely perform ordinary data preparation and manipulation incidental to the abstract analysis. Considered individually and as an ordered combination, the claims do not recite a non-generic technological arrangement or improvement to computer functionality.
With respect to independent claims 8 and 15, the recited processor, memory, computer program and storage medium merely implement substantially the same abstract process using generic computer components and do not provide a technological improvement. However, Applicant’s amendment of claim 15 to recite a “non-transitory computer-readable storage medium” overcomes the previously stated Step 1 rejection concerning the term “non-transient.” The Step 1 rejection of claim 15 is therefore withdrawn, while the Step 2A and Step 2B rejection is maintained.
Accordingly, Applicant’s arguments are not persuasive and the rejection of the pending claims under 35 U.S.C. § 101 is maintained.
Regarding Applicant’s arguments against the 102/103 rejections on p.22 – 32: Applicant’s arguments have been fully considered but are not persuasive. To the extent the amendments incorporate limitations not previously relied upon in the anticipation rejection, the rejection has been modified as set forth above to account for the amended claim language.
With respect to the rejection over Bin in view of Yang, Applicant argues that the references concern different fields and that there would have been no motivation to combine their teachings. This argument is unpersuasive because the rejection does not require bodily incorporation of Yang’s image-processing system into Bin. Rather, Yang is relied upon for its teaching of using probability-transition information and distances between data points to characterize similarity. Both references concern analyzing relationships between data represented as nodes/features, and one of ordinary skill would have recognized the applicability of Yang’s known similarity-analysis technique to Bin’s network data for providing a quantitative determination of similarity between data associated with nodes. The fact that Yang applies the technique to image features does not, by itself, negate the reason to use the known mathematical technique for analogous data-analysis purposes.
Applicant further argues that Yang does not disclose the claimed equations verbatim. However, obviousness does not require that a secondary reference expressly disclose the claimed subject matter in precisely the same form. Yang teaches the underlying use of transition probabilities, distances, and a probability transition matrix for determining similarity between data points. The rejection relies upon these teachings in combination with Bin, rather than upon Yang as an anticipation reference. Accordingly, Applicant’s arguments directed to what Bin or Yang individually fails to disclose does not address the combined teachings of the references and the rationale provided in the rejection.
With respect to Rausch, Applicant argues that Rausch concerns ETL/data warehousing rather than social-network prediction. This argument is likewise unpersuasive. Rausch is relied upon for its teachings concerning transforming, structuring, and processing data, including database-related processing, rather than for the entirety of the claimed social-network prediction method. One of ordinary skill would have recognized that such known data-processing techniques may predictably be employed to prepare and structure data for subsequent analytical processing. The references need not address identical problems or fields where there is an articulated reason to apply the known technique to the primary reference.
Finally, Applicant’s arguments concerning Finkbeiner are also unpersuasive because Finkbeiner is relied upon for the claimed mathematical/normalization technique, not for its particular neurological application. The mere fact that a mathematical relationship is disclosed in a different application does not establish that its use for the claimed data analysis would have been nonobvious where the rejection provides a reason for employing the known mathematical technique and the combination would have yielded predictable results.
Accordingly, Applicant’s arguments concerning the individual purposes and fields of the cited references do not overcome the rejection based upon their combined teachings. The rejection under 35 U.S.C. § 103 are therefore maintained for the reasons set forth above.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1, 4 – 8, 11 – 15, and 18 – 20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more, and therefore des not recite patent-eligible subject matter.
Step 2A Prong 1: The abstract idea is defined by the elements of:
acquiring original data of training users, and performing training user information division on the original data to obtain target information;
defining relationships between the training users according to the target information to obtain a user relationship network;
performing node feature extraction on the user relationship network… environmental influences of users; and
constructing a user relationship analysis model based on the feature data, and performing relationship prediction … obtain user relationships of the user to be tested;
transforming the original data to obtain standard data;
calculating a neighborhood distance of the standard data to obtain a neighborhood mean value;
obtaining a weight coefficient of the standard data, calculating matrix elements of the … and generating a probability transition matrix according to the matrix elements:
processing abnormal information of the standard data according to the similarity value to obtain the target information;
wherein the transforming the original data into standard data comprises:
converting the original data into a database according to a preset file template to obtain a simulation output file;
generating a database object set according to the simulation output file; and
performing file parsing on the database object set to obtain the standard data.
These limitations describe a data-analysis system for predicting relationships between users in a social network, in which these operations include collecting data, analyzing data, generating models, and predicting relationships among users. Each of these steps constitutes data observation, evaluation, organization, modeling, and prediction—activities that can be performed in the human mind or with pen and paper. For example, acquiring original user data merely involves gathering information; dividing user information involves categorizing or organizing information; defining relationships between users involves evaluating social relationships based on known facts; extracting features according to influencing factors involves identifying and selecting characteristics deemed relevant; constructing a relationship analysis model involves formulating a conceptual or mathematical model; and predicting user relationships involves drawing conclusions based on that model. These are mental processes that can be performed mentally or with pen-and-paper.
Step 2A Prong 2: For independent claim 1, the claims do not integrate an abstract idea into a practical application. Claim 1 recites no additional elements beyond the abstract idea itself. Because Claim 1 recites no additional elements beyond the mental processing steps, the claim does not integrate the judicial exception into a practical application. Furthermore, claim 1 recites no machine at all. There is no processor, memory, or computer. The steps are conceptual and can be performed mentally, which weighs heavily against eligibility.
Step 2A Prong 2: For independent claim 8 and 15, although claims 8 and 15 recite an electronic apparatus and a computer-readable storage medium, respectively, these claims merely implement the same abstract method of claim 1 using generic computer components such as a processor, memory, and executable instructions. The use of such generic computer elements to carry out abstract mental processes does not integrate the judicial exception into a practical application. These additional limitations do not improve a computer or other technology, do not require a particular machine, and do not impose a meaningful restriction of the abstract idea.
Step 2B: For independent claim 1, because claim 1 is directed to a judicial exception and does not integrate the exception into a practical application, the analysis proceeds to Step 2B to determine whether the claim includes additional elements that amount to significantly more than the abstract idea. Claim 1 does not include any such additional elements. As discussed above, claim 1 recites only abstract mental processes involving the acquisition, organization, analysis, modeling, and prediction of user relationship information, and does not recite any machine, processor, memory, or specialized hardware. The claim therefore amounts to nothing more than the abstract idea itself.
Step 2B: For independent claims 8 and 15, the additional elements recited therein are limited to generic computer components, such as processor, a memory, and computer-executable instructions stored on a non-transitory computer-readable medium. These components are described at a high level of generality and are used only to perform routine functions of executing instructions, storing data, and processing information. The use of such generic computer components to implement an abstract mental process does not constitute an inventive concept. The claims do not recite any non-generic computer architecture, specialized processing technique, or improvement to the functioning of the computer itself, but instead merely apply the abstract idea.
For dependent claims 2 – 7, 9 – 14, and 16 – 20, these claims cover or fall under the same abstract idea of a mental processes. The dependent claims further recite additional data processing steps, mathematical calculations, probability and similarity determinations, feature extraction operations, and data formatting or parsing techniques. These limitations merely add further detail to the abstract data analysis and mathematical modeling already recited in the independent claims. For instance:
Claims 4, 11 and 18 is directed to mental processes and organizing human activity by judging relationships between users and labeling users based on those judgements, as the claims recite additional social-relationship determinations and graph construction steps, all conceptual.
Claims 5, 12 and 19 is directed to mental processes involving feature identification and weighting, as the claims recite further data-analysis and feature-engineering steps, which are abstract and do not implement a technical improvement.
Claims 6, 13, and 20 recite additional mathematical operations, expressly treated as an abstract idea under §101. Further, Examiner does not find the presence of these two abstract ideas in the claims render the claims non-abstract, see MPEP 2106.04.I discussing Recognicorp (stating combining “one abstract idea (math) to another abstract idea (encoding and decoding) does not render the claim non-abstract”).
Claims 7 and 14 is directed to mental processes involving prediction and evaluation, as the claims recite data analysis. Predicting social relationships without technological impact does not integrate the abstract idea.
Step 2A Prong 2 and Step 2B: For dependent claims, with respect to the dependent claims, none integrates the judicial exception into a practical application. Instead, each dependent claim merely adds further detail to the abstract mental processes and mathematical analysis already recited in independent claim 1, without improving a computer or other technology, without requiring a particular machine, and without effecting a transformation of matter.
Claims 4, 11 and 18 further recite performing relationship judgment, labeling users, generating an initial user relationship network, and defining relationships within that network. These limitations are directed to conceptual evaluation and categorization of social relationships, which constitute mental processes and methods of organizing human activity. The claims do not impose any technological limitation or recite a specific technical mechanism for performing these judgments. Therefore, claims 4, 11 and 18 do not integrate the abstract idea into a practical application.
Claims 5, 12 and 19 further recite establishing an influencing factor function and performing feature extraction using that function. These limitations amount to defining and applying mathematical or logical rules to determine how certain factors influence user relationships. Such feature weighting and evaluation are abstract analytical techniques and mental processes. The claims do not recite a technological improvement or a specific implementation that improves computing technology. Accordingly, claims 5, 12 and 19 do not integrate the abstract idea into a practical application.
Claims 6, 13 and 20 further recite constructing the relationship analysis model using the mathematical formula H = (S – T)/(S + T). This limitation is a pure mathematical relationship used to process data. The claims do not apply the formula in a technological environment or to improve a technical process, but instead use it solely to generate an informational relationship value. As such, claim 6, 13 and 20 do not integrate the abstract into a practical application.
Claims 7 and 14 further recite extracting feature data for a user to be tested, calculating a relationship value, and determining a user relationship based on that value. These limitations describe abstract prediction and evaluation of information using a conceptual model. The claims do not recite any technological application of the prediction or any improvement to computer functionality. Accordingly, claims 7 and 14 do not integrate the abstract idea into a practical application.
Additionally, these elements and their limitations are “merely indicating a field of use or technological environment in which to apply a judicial exception do not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application” (MPEP 2106.05(h)). Accordingly, these additional elements do not integrate the abstract idea into a practical application because it does not impose any meaningful limits on practicing the abstract idea.
Therefore, claims 1, 4 – 8, 11 – 15, and 18 – 20 are rejected under 35 U.S.C. § 101 for being directed to an abstract idea without sufficient integration into a practical application, and the additional elements do not add significantly more than the judicial exception
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 – 8, 11 – 15, and 18 – 20 are rejected under 35 U.S.C. 103 as being unpatentable over Bin (CN103795613 A, machine translation via Espacenet) in view of Yang (US9342991 B2) in further view of Rausch (US20070174308 A1).
Regarding claims 1 and 15:
Bin discloses:
acquiring original data of training users, and performing training user information division on the original data to obtain target information; [¶0013]: The disclosure is able to select and view users within a social network as well as access data regarding a user. Additionally, [¶0014]: A user’s relationship data is then processed to predict the relationship between the user and other users within the test data. [¶0031]: “…features are divided into friend relationships and non-friend relationships according to the established friend relationship model”
defining relationships between the training users according to the target information to obtain a user relationship network; [¶0014]: Testing is conducted on a user’s relationship chart and data in order to further analyze a user’s relationship with other users in the test data.
performing node feature extraction on the user relationship network according to preset influencing factors to obtain feature data corresponding to the influencing factors, wherein the preset influencing factors comprise at least two selected from the group consisting of: personal interest reflecting intrinsic attributes of users, friend relationships reflecting social connections of users, and community drive reflecting environmental influences of users; and [¶0047 - 0048] Features are extracted based on collected data in order to establish relationships between users. [¶0053] Nodes are identified within the social network and are then categorized according to whether the nodes are directly connected or not.
constructing a user relationship analysis model based on the feature data, and [Fig. 2; ¶0047] Features are extracted based on collected data in order to establish relationships between users.
performing relationship prediction on user data of a preset user to be tested using the user relationship analysis model to obtain user relationships of the user to be tested. [¶0014]: A user’s relationship data is then processed to predict the relationship between the user and other users within the test data.
transforming the original data to obtain standard data [¶0047] Once node feature extracted has been conducted, the information is standardized to a format that is able to be read;
calculating a neighborhood distance of the standard data to obtain a neighborhood mean value [¶0055] The social network is used as a basis for the calculation method. The distance between the networks of users and the nodes are calculated using different formulas/algorithms, i.e., Dijkstra algorithm or Floyd algorithm;
Though Bin discloses a calculation method used to determine distances between users within a topological network, Bin does not explicitly disclose a probability transition matrix. Thus, Yang teaches:
obtaining a weight coefficient of the standard data, calculating matrix elements of the standard data according to the neighborhood mean value and the weight coefficient using the following formula, and generating a probability transition matrix according to the matrix elements: p=zMeanid(xi,xik),pi>0, wherein, p represents the probability transition matrix; Mean represents the neighborhood mean value of an i-th piece of standard data; z represents the weight coefficient of the standard data; d(xi,xik) represents a distance between the i-th piece of standard data and a neighborhood xik of the i-th piece of standard data xi; [col. 5 line 11]:
PNG
media_image1.png
130
381
media_image1.png
Greyscale
The equation relates directly to Markov’s chains, the formal mathematical representation of the probabilities in a Markov chain—the fundamentals of the equation is to describe the likelihood of moving from one state to another in one step, with each row summing to 1.
[Examiner’s Note: The reference teaches computing a diffusion distance between two nodes by aggregating transition probability differences across a neighborhood of nodes and normalizing by a stationary distribution term. This computation reflects a neighborhood-based aggregate measure of similarity that incorporates local connectivity information and density weighting. Although the reference expresses this aggregation using squared transition probability differences, it nevertheless teaches the structural concept of computing similarity as a normalized neighborhood aggregation of pairwise transition behavior. A person of ordinary skill in the art would recognize that computing a neighborhood mean over a specific pairwise measure, as recited in the claim, represents a predictable mathematical variation of the diffusion-based aggregation technique disclosed in the reference.)
describing a similarity of the standard data based on the probability transition matrix to obtain a similarity value using the following formula: f=∑i=1Tpz,x, i, wherein, f represents the similarity value; T represents a preset similarity parameter; p represents the probability transition matrix; z represents the weight coefficient of the standard data; xi indicates as the i-th piece of standard data; and [¶0030]: The disclosure teaches creating similarity matrixes being computed by comparing distributions using the Kullback-Leibler divergence. The reference teaches computing a similarity (diffusion distance) between two data points using entries of a probability transition matrix p(t)iq . (Examiner’s Note: Both formulas aggregate transition probability terms, produce a scalar similarity metric, as well as they operate in the same probability-transition space.)
processing abnormal information of the standard data according to the similarity value to obtain the target information. [Fig. 6; ¶0049]: The similarity matrix is normalized, which then allows for the data to generate a Markov transition matrix.
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date to modify Bin’s disclosed method for predicting friend relationships within an online social network with that of transition matrix formulas as well as similarity matrix formulas, as taught by Yang, in order to accurately and efficiently produce and process a similarity score between neighboring nodes in a network.
Neither Bin nor Yang teach the limitations below. Thus, Rausch teaches:
converting the original data into a database according to a preset file template to obtain a simulation output file [Fig. 4; ¶0030]: The disclosure teaches a process to convert metadata objects into executable code segments. Additionally, [Fig. 2; ¶0022]: teaches a database that is coupled to the plethora of workstations, applications and networks;
generating a database object set according to the simulation output file [Fig. 2; ¶0022]: teaches a database that is coupled to the plethora of workstations, applications and networks; and
performing file parsing on the database object set to obtain the standard data [¶0037]: Transformation templates are able to be exported and shared; this process is conducted through parsing appropriate XML code;
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date to modify Bin’s disclosed method for predicting friend relationships within an online social network and Yang’s disclosed transition matrix formulas as well as similarity matrix formulas with performing file parsing as well as generating a database, as taught by Rausch, as it would’ve been applying a known technique of similarity matrix formulas to the environment of an online social network.
Regarding claims 4 and 18:
Bin discloses:
performing relationship judgment on the training users according to the target information, and labeling the training users according to a result of the relationship judgment to obtain an initial user set; [¶0047]: The system measures the content of information between users within a network and characterizes the relationship between user nodes. Additionally, [¶0067]: teaches allowing the user to select specific relationship data and conduct a ‘relationship prediction’ between chosen users.
generating an initial user relationship network of the training users according to the initial user set; and [¶0053]: A social network is defined and a complete graph is generated to display an initial set of users. If they are friends, they are connected; if users are recognized not to be friends, they are labeled as “non-friends”.
defining relationships of the initial user relationship network to obtain a user relationship network. [¶0030 - 0033]: Based on collected data (i.e., user check-in time, location, type, and user’s friend relationships), a friend prediction model is generated and nodes are assigned accordingly.
Regarding claims 5 and 19:
Bin discloses:
performing information extraction on the user relationship network according to the influencing factors to obtain a feature set corresponding to the influencing factors; [¶0047]: Features are extracted and selected from the collected data (e.g., social topology, user check-in location type, and user check-in location).
establishing an influencing factor function according to the feature set; [¶0052 – 0053]: Based on the features collected, a social network is defined by function “G s (U s , E s ), node.” [Examiner’s Note: With respect to BRI, the “influencing factor function” is interpreted as a formula or rule that decides how much each factor “influences” a user relationship. The Bin reference uses different social signals (user social topology, check-in data) as influencing factors and performs friend-relationship prediction over that network.]
performing feature extraction on the user relationship network using the influencing factor function to obtain the feature data corresponding to the influencing factors. [¶0055 - 0056]: Once the social network is defined, friend edges and distances are defined within the topology network.
Regarding claims 7 and 14:
Bin discloses:
performing feature extraction on the user data to obtain feature data of the user to be tested; [¶0047]: Features are extracted based on the collected data to then be used to characterize relationships between the user nodes.
performing a relationship calculation on the feature data of the user to be tested using the user relationship analysis model to obtain a relationship value; and [¶0047 - 0048]: The disclosure uses the information extracted in order to process classification algorithms to measure information content of the selected features, thus, producing different parameter frequencies.
determining the user relationship of the user to be tested according to the relationship value. [¶0053]: A social network is defined and friend edges are created between the nodes, distinguishing between users that are friends and non-friends.
Regarding claim 8:
Bin discloses:
acquiring original data of training users, and performing training user information division on the original data to obtain target information; [¶0013]: The disclosure is able to select and view users within a social network as well as access data regarding a user. Additionally, [¶0014]: A user’s relationship data is then processed to predict the relationship between the user and other users within the test data. [¶0031]: “…features are divided into friend relationships and non-friend relationships according to the established friend relationship model”
defining relationships between the training users according to the target information to obtain a user relationship network; [¶0014]: Testing is conducted on a user’s relationship chart and data in order to further analyze a user’s relationship with other users in the test data.
performing node feature extraction on the user relationship network according to preset influencing factors to obtain feature data corresponding to the influencing factors, wherein the preset influencing factors comprise at least two selected from the group consisting of: personal interest reflecting intrinsic attributes of users, friend relationships reflecting social connections of users, and community drive reflecting environmental influences of users; and [¶0047 - 0048] Features are extracted based on collected data in order to establish relationships between users. [¶0053] Nodes are identified within the social network and are then categorized according to whether the nodes are directly connected or not.
constructing a user relationship analysis model based on the feature data, and [Fig. 2; ¶0047] Features are extracted based on collected data in order to establish relationships between users.
performing relationship prediction on user data of a preset user to be tested using the user relationship analysis model to obtain user relationships of the user to be tested. [¶0014]: A user’s relationship data is then processed to predict the relationship between the user and other users within the test data.
transforming the original data to obtain standard data [¶0047] Once node feature extracted has been conducted, the information is standardized to a format that is able to be read;
calculating a neighborhood distance of the standard data to obtain a neighborhood mean value [¶0055] The social network is used as a basis for the calculation method. The distance between the networks of users and the nodes are calculated using different formulas/algorithms, i.e., Dijkstra algorithm or Floyd algorithm;
Though Bin discloses a calculation method used to determine distances between users within a topological network, Bin does not explicitly disclose a probability transition matrix. Thus, Yang teaches:
at least one processor; and [col. 12, line 62]: The disclosure teaches one or more processors (CPUs).
obtaining a weight coefficient of the standard data, calculating matrix elements of the standard data according to the neighborhood mean value and the weight coefficient using the following formula, and generating a probability transition matrix according to the matrix elements: p=zMeanid(xi,xik),pi>0, wherein, p represents the probability transition matrix; Mean represents the neighborhood mean value of an i-th piece of standard data; z represents the weight coefficient of the standard data; d(xi,xik) represents a distance between the i-th piece of standard data and a neighborhood xik of the i-th piece of standard data xi; [col. 5 line 11]:
PNG
media_image1.png
130
381
media_image1.png
Greyscale
The equation relates directly to Markov’s chains, the formal mathematical representation of the probabilities in a Markov chain—the fundamentals of the equation is to describe the likelihood of moving from one state to another in one step, with each row summing to 1.
[Examiner’s Note: The reference teaches computing a diffusion distance between two nodes by aggregating transition probability differences across a neighborhood of nodes and normalizing by a stationary distribution term. This computation reflects a neighborhood-based aggregate measure of similarity that incorporates local connectivity information and density weighting. Although the reference expresses this aggregation using squared transition probability differences, it nevertheless teaches the structural concept of computing similarity as a normalized neighborhood aggregation of pairwise transition behavior. A person of ordinary skill in the art would recognize that computing a neighborhood mean over a specific pairwise measure, as recited in the claim, represents a predictable mathematical variation of the diffusion-based aggregation technique disclosed in the reference.)
describing a similarity of the standard data based on the probability transition matrix to obtain a similarity value using the following formula: f=∑i=1Tpz,x, i, wherein, f represents the similarity value; T represents a preset similarity parameter; p represents the probability transition matrix; z represents the weight coefficient of the standard data; xi indicates as the i-th piece of standard data; and [¶0030]: The disclosure teaches creating similarity matrixes being computed by comparing distributions using the Kullback-Leibler divergence. The reference teaches computing a similarity (diffusion distance) between two data points using entries of a probability transition matrix p(t)iq . (Examiner’s Note: Both formulas aggregate transition probability terms, produce a scalar similarity metric, as well as they operate in the same probability-transition space.)
processing abnormal information of the standard data according to the similarity value to obtain the target information. [Fig. 6; ¶0049]: The similarity matrix is normalized, which then allows for the data to generate a Markov transition matrix.
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date to modify Bin’s disclosed method for predicting friend relationships within an online social network with that of transition matrix formulas as well as similarity matrix formulas, as taught by Yang, in order to accurately and efficiently produce and process a similarity score between neighboring nodes in a network.
Neither Bin nor Yang teach the limitations below. Thus, Rausch teaches:
converting the original data into a database according to a preset file template to obtain a simulation output file [Fig. 4; ¶0030]: The disclosure teaches a process to convert metadata objects into executable code segments. Additionally, [Fig. 2; ¶0022]: teaches a database that is coupled to the plethora of workstations, applications and networks;
generating a database object set according to the simulation output file [Fig. 2; ¶0022]: teaches a database that is coupled to the plethora of workstations, applications and networks; and
performing file parsing on the database object set to obtain the standard data [¶0037]: Transformation templates are able to be exported and shared; this process is conducted through parsing appropriate XML code;
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date to modify Bin’s disclosed method for predicting friend relationships within an online social network and Yang’s disclosed transition matrix formulas as well as similarity matrix formulas with performing file parsing as well as generating a database, as taught by Rausch, as it would’ve been applying a known technique of similarity matrix formulas to the environment of an online social network.
Regarding claim 11:
Bin discloses:
performing relationship judgment on the training users according to the target information, and labeling the training users according to a result of the relationship judgment to obtain an initial user set; [¶0047]: The system measures the content of information between users within a network and characterizes the relationship between user nodes. Additionally, [¶0067]: teaches allowing the user to select specific relationship data and conduct a ‘relationship prediction’ between chosen users.
generating an initial user relationship network of the training users according to the initial user set; and [¶0053]: A social network is defined and a complete graph is generated to display an initial set of users. If they are friends, they are connected; if users are recognized not to be friends, they are labeled as “non-friends”.
defining relationships of the initial user relationship network to obtain a user relationship network. [¶0030 - 0033]: Based on collected data (i.e., user check-in time, location, type, and user’s friend relationships), a friend prediction model is generated and nodes are assigned accordingly.
Regarding claim 12:
Bin discloses:
performing relationship judgment on the training users according to the target information, and labeling the training users according to a result of the relationship judgment to obtain an initial user set; [¶0047]: The system measures the content of information between users within a network and characterizes the relationship between user nodes. Additionally, [¶0067]: teaches allowing the user to select specific relationship data and conduct a ‘relationship prediction’ between chosen users.
generating an initial user relationship network of the training users according to the initial user set; and [¶0053]: A social network is defined and a complete graph is generated to display an initial set of users. If they are friends, they are connected; if users are recognized not to be friends, they are labeled as “non-friends”.
defining relationships of the initial user relationship network to obtain a user relationship network. [¶0030 - 0033]: Based on collected data (i.e., user check-in time, location, type, and user’s friend relationships), a friend prediction model is generated and nodes are assigned accordingly.
Claims 6, 13 and 20 are rejected under 35 U.S.C. § 103 as being unpatentable over Bin (CN103795613 A, machine translation via Espacenet) in view of Yang (US9342991 B2) in further view of Finkbeiner (US9359445 B2)
Regarding claims 6, 13 and 20:
Though Bin discloses constructing a user relationship analysis model based on feature data [¶0014], neither Bin nor Yang disclose using the formula below. Thus, Finkbeiner teaches:
constructing the user relationship analysis model based on the feature data using the following formula: H=S-T/S+T; wherein, H represents the user relationship analysis model; S represents the feature data of the training user; T represents a preset relationship parameter of the training user. [col. 49 line 4]: “normalizing by difference-over-sum as Sr−St/Sr+St=Ci. Ci=1 indicates ideal colocalization, Ci=0 no colocalization above chance, and Ci<0 anti-localization”
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date to apply the normalized difference formulation taught by Finkbeiner to the feature data produced by the Bin/Yang combination in order to generate a normalized relationship score. Applying a known normalization operator to known feature quantities to produce a bounded relationship score constitutes a predictable use of prior art elements according to their established functions. Under KSR, selecting one known normalization formula over another for computing a comparison metric is a routine design choice.
Pertinent Art
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Bill Chen whose telephone number is (571)270-0660. The examiner can normally be reached Monday - Friday 8:30am - 5:00pm.
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
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, Nathan Uber can be reached on (571) 270-3923. 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.
/BILL CHEN/Examiner, Art Unit 3626
/NATHAN C UBER/Supervisory Patent Examiner, Art Unit 3626
Prosecution Insights