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 .
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 1-15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Step 1: Claims 2-12 are method claims. Claims 1, 13-15 are machine/system/product claims. Therefore, claims 1-15 are directed to either a process, machine, manufacture or composition of matter.
With respect to claim 1:
Step 2A – Prong 1:
…
…
generating, based on the TKGs stored in the database, relationship vectors that describe the relations for each node of the TGKs for each available timestep; (mental process – a person can manually generate relationship vectors with the assistance of a pen/paper.)
using the generated relationship vectors to create a sequential dataset including at least one vector sequence set for each node of the TKGs; (mental process – a person can manually use the generated relationship vectors to create a sequential dataset with the assistance of a pen/paper.)
using the at least one vector sequence sets as sequential input samples for training and execution of a pattern model that is learned to predicts for one or more future timesteps of interest, future relations for each node of the TKGs; (mental process – a person can manually predict future relations for each node of the TKGs with the assistance of a pen/paper.)
…
using the predicted future relations and nodes to assemble predicted future TKGs describing a crime related scenario in an area of interest per future time steps of interest; (mental process – a person can manually use the predicted future relations and nodes to assemble predicted future TKGs describing a crime related scenario in an area of interest per future time steps of interest with the assistance of a pen/paper.)
a forecasting-based action recommendation system configured to iteratively compute one or more actions acting on the predicted future crime related scenario that steer the predicted future crime related scenario towards a desired scenario; (mental process – a person can manually iteratively compute one or more actions acting on the predicted future crime related scenario that steer the predicted future crime related scenario towards a desired scenario with the assistance of a pen/paper.)
and control means configured to automatically adapt monitoring and/or surveillance devices deployed in the area of interest based on the computed one or more actions. (mental process – a person can manually adapt monitoring and/or surveillance devices deployed in the area of interest with the assistance of a pen/paper.)
Step 2A – Prong 2: This judicial exception is not integrated into a practical application.
A predictive policing system, the system comprising: a database comprising crime related scenario states in a number of past timesteps represented as temporal knowledge graphs (TKGs); (Adding insignificant extra-solution activity to the judicial exception - see MPEP 2106.05(g)).
one or more crime prediction processing devices that, alone or in combination, are configured to provide for execution of the following steps: (mere instructions to apply the exception using a generic computer component – processing devices applies exception.)
…
…
…
training and execution of a forecasting model that is learned to predict, for one or more future timesteps of interests the nodes of the TKG associated with each of the predicted future relations; Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: High level recitation of training the machine learning engine to predict, for one or more future timesteps of interests the nodes of the TKG associated with each of the predicted future relations.);
Step 2B: The claim does not include additional elements considered individually and in combination that are sufficient to amount to significantly more than the judicial exception.
A predictive policing system, the system comprising: a database comprising crime related scenario states in a number of past timesteps represented as temporal knowledge graphs (TKGs); (MPEP 2106.05(d)(II) indicate that merely “ Storing and retrieving information in memory” is a well‐understood, routine, conventional function when it is claimed in a merely generic manner (as it is in the present claim – the crime related scenario states in a number of past timesteps represented as temporal knowledge graphs are merely stored in a database). Thereby, a conclusion that the claimed distribute step is well-understood, routine, conventional activity is supported under Berkheimer.)
one or more crime prediction processing devices that, alone or in combination, are configured to provide for execution of the following steps: (mere instructions to apply the exception using a generic computer component – processing devices applies exception.)
…
…
…
training and execution of a forecasting model that is learned to predict, for one or more future timesteps of interests the nodes of the TKG associated with each of the predicted future relations; (Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: High level recitation of training the machine learning engine to predict, for one or more future timesteps of interests the nodes of the TKG associated with each of the predicted future relations.);
Claim 2 is substantially similar to claim 1, but has the following additional elements:
With respect to claim 2:
Step 2A – Prong 1:
The substantially similar additional elements from claim 1 contain mental processes (see claim 1)
Step 2A – Prong 2: This judicial exception is not integrated into a practical application.
A computer-implemented method for Temporal Knowledge Graph (TKG) forecasting, the method comprising: (mere instructions to apply the exception using a generic computer component – computer applies exception.)
… by a relation prediction module based on a graph database comprising scenario states in a number of past timesteps represented as TKGs, … (mere instructions to apply the exception using a generic computer component – relation prediction module applies exception.)
With respect to claim 3:
Step 2A – Prong 1:
The method according to claim 2, further comprising: computing, by the entity prediction module, a time-dependent graph embedding for each relation contained in the TKGs; (mental process – a person can manually compute a time-dependent graph embedding for each relation contained in the TKGs with the assistance of a pen/paper.)
and computing, by the entity prediction module based on a predefined similarity metric, a similarity of each relation contained in the TKGs to each of the other relations contained in the TKGs and creating, based on the computed similarities, relation similarity matrices for each relation and for each timestep. (mental process – a person can manually compute a similarity of each relation contained in the TKGs to each of the other relations contained in the TKGs and creating, based on the computed similarities, relation similarity matrices for each relation and for each timestep with the assistance of a pen/paper.)
With respect to claim 4:
Step 2A – Prong 1:
The method according to claim 3, wherein the graph embedding for the relationships is computed on the graph database including scenario states of all available past timesteps. (mental process – a person can recognize that the graph embedding for the relationships is computed on the graph database including scenario states of all available past timesteps.)
With respect to claim 5:
Step 2A – Prong 1:
The method according to claim 3, (mental process from claim 3)
Step 2A – Prong 2: This judicial exception is not integrated into a practical application.
further comprising: using the relation similarity matrices as sequential input samples for training and execution of the forecasting model that is learned to predict, based on past similarity matrices for a relation, future similarity matrices for the respective relation. Adding the words “apply it” (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea - see MPEP 2106.05(f) – Examiner’s note: High level recitation of training the machine learning engine to predict future similarity matrices for the respective relation.);
With respect to claim 6:
Step 2A – Prong 1:
The method according to claim 2, wherein the forecasting model is a sequential model implemented in form of a recurrent neural network (RNN). (mental process – a person can recognize that the forecasting model is a sequential model implemented in form of a recurrent neural network.)
With respect to claim 7:
Step 2A – Prong 1:
The method according to claim 5, further comprising: extracting, from the predicted future similarity matrices, for each predicted future relation an entity with a highest predicted similarity and associating the extracted entity with the respective [subject, relation] or [relation, object] pair to obtain all predicted triples of future TKGs for timesteps of interest. (mental process – a person can manually extract an entity with a highest predicted similarity and associating the extracted entity with the respective [subject, relation] or [relation, object] pair to obtain all predicted triples of future TKGs for timesteps of interest with the assistance of a pen/paper.)
With respect to claim 8:
Step 2A – Prong 1:
The method according to claim 7, wherein the extraction of a predicted entity for a predicted future relation takes into consideration the neighborhood of the respective entity across all timesteps by means of a neighborhood aggregation matrix created by computing, and subsequently aggregating over, the similarity of each relation in the neighborhood of the predicted entity across all timesteps to the respective predicted future relation. (mental process – a person can recognize that the extraction of a predicted entity for a predicted future relation takes into consideration the neighborhood of the respective entity across all timesteps by means of a neighborhood aggregation matrix created by computing, and subsequently aggregating over, the similarity of each relation in the neighborhood of the predicted entity across all timesteps to the respective predicted future relation.)
With respect to claim 9:
Step 2A – Prong 1:
The method according to claim 2, further comprising: encoding the generated relationship vectors for dimensionality reduction prior to creating the sequential dataset, and/or encoding the relation similarity matrices for dimensionality reduction prior to providing the relation similarity matrices to the forecasting model. (mental process – a person can manually encode the generated relationship vectors for dimensionality reduction prior to creating the sequential dataset with the assistance of a pen/paper.)
With respect to claim 10:
Step 2A – Prong 1:
The method according to claim 2, further comprising: embedding the relation prediction module and the entity prediction module in a forecasting-based action recommendation system that iteratively computes appropriate actions acting on a predicted scenario that steer the predicted scenario towards a desired scenario. (mental process – a person can manually embed the relation prediction module and the entity prediction module in a forecasting-based action recommendation system that iteratively computes appropriate actions acting on a predicted scenario that steer the predicted scenario towards a desired scenario with the assistance of a pen/paper.)
With respect to claim 11:
Step 2A – Prong 1:
The method according to claim 2, wherein the graph database is a database containing crime related scenario states in a number of past timesteps represented as TKGs, the method further comprising: predicting future TKGs describing a crime related scenario in an area of interest per future time steps of interest, (mental process – a person can manually predict future TKGs describing a crime related scenario in an area of interest per future time steps of interest with the assistance of a pen/paper.)
iteratively computing one or more actions acting on the predicted future crime related scenario that steer the predicted future crime related scenario towards a desired scenario, (mental process – a person can manually iteratively compute one or more actions acting on the predicted future crime related scenario that steer the predicted future crime related scenario towards a desired scenario with the assistance of a pen/paper.)
and automatically adapting monitoring and/or surveillance devices deployed in the area of interest based on the computed one or more actions. (mental process – a person can manually adapting monitoring and/or surveillance devices deployed in the area of interest based on the computed one or more actions with the assistance of a pen/paper.)
With respect to claim 12:
Step 2A – Prong 1:
The method according to claim 2, wherein the graph database is a database of a public health related infection control service containing infection related scenario states in a number of past timesteps represented as TKGs, the method further comprising: predicting future TKGs describing infection development on a predefined geographical level per future time steps of interest, (mental process – a person can manually predict future TKGs describing a crime related scenario in an area of interest per future time steps of interest with the assistance of a pen/paper.)
iteratively computing one or more actions acting on the predicted future infection related scenario that steer the predicted future scenario towards a desired scenario, (mental process – a person can manually iteratively compute one or more actions acting on the predicted future crime related scenario that steer the predicted future crime related scenario towards a desired scenario with the assistance of a pen/paper.)
and automatically adapting access control devices for public buildings, advertising on digital advertising panels, and/or frequency of public transport based on the computed one or more actions. (mental process – a person can manually adapting monitoring and/or surveillance devices deployed in the area of interest based on the computed one or more actions with the assistance of a pen/paper.)
Claim 13 is substantially similar to claim 1, but has the following additional elements:
With respect to claim 13:
Step 2A – Prong 1:
A processing system comprising one or more processors which, alone or in combination, are configured to provide for execution of a method for Temporal Knowledge Graph (TKG) forecasting, the method comprising: (mere instructions to apply the exception using a generic computer component – processing system applies exception.)
… by a relation prediction module based on a graph database comprising scenario states in a number of past timesteps represented as TKGs, … (mere instructions to apply the exception using a generic computer component – relation prediction module applies exception.)
Claim 14 is rejected on the same grounds under 35 U.S.C. 101 as claim 1 as they are substantially similar. Mutatis mutandis.
Claim 15 is substantially similar to claim 1, but has the following additional elements:
With respect to claim 15:
Step 2A – Prong 1:
A tangible, non-transitory computer-readable medium having instructions thereon which, upon being executed by one or more processors, alone or in combination, provide for execution of a method of Temporal Knowledge Graph (TKG) forecasting, the method comprising: (mere instructions to apply the exception using a generic computer component – tangible, non-transitory computer-readable medium module applies exception.)
… by a relation prediction module based on a graph databased comprising scenario states in a number of past timesteps represented as TKGs, … (mere instructions to apply the exception using a generic computer component – relation prediction module applies exception.)
… by an entity prediction module, … (mere instructions to apply the exception using a generic computer component – entity prediction module applies exception.)
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 1-15 are rejected under 35 U.S.C. 103 as being unpatentable over Jin et al. (“Recurrent Event Network: Autoregressive Structure Inference over Temporal Knowledge Graphs”) hereinafter known as Jin in view of Jacobs et al. (“ProcK: Machine Learning for Knowledge-Intensive Processes”) hereinafter known as Jacobs in view of Muijs (EP3391352B1) hereinafter known as Muijs.
Regarding independent claim 1, Jin teaches:
A predictive policing system, the system comprising: a database comprising crime related scenario states in a number of past timesteps represented as temporal knowledge graphs (TKGs); (Jin [Page 1, Col. 1, Paragraph 2]: “each fact may not be true forever and hence it is useful to associate each fact with a timestamp as a constraint, yielding a temporal knowledge graph (TKG)” Jin teaches that there is a collection/database of past timestamps associated with a fact, creating a TKG.)
…
generating, based on the TKGs stored in the database, relationship vectors that describe the relations for each node of the TGKs for each available timestep; (Jin [Page 1, Col. 1, Paragraph 1]: “RE-NET employs a recurrent event encoder to encode past facts, and uses a neighborhood aggregator to model the connection of facts at the same timestamp” [Page 3, Col. 2, Paragraph 3]: “where e_{s}, e_{r} in R^{d} are learnable embedding vectors specified for subject entity s and relation r.” Jin teaches that the RE-NET encodes connections of facts at a given timestamp and are vectors detailing the relation.)
using the generated relationship vectors to create a sequential dataset including at least one vector sequence set for each node of the TKGs; (Jin [Page 2, Col. 2, Paragraph 4]: “Formally, we represent TKGs as sequences, and then build an autoregressive generative model on the sequences” Jin teaches building the TKGs as sequences and subsequently as a generative model based on the sequences.)
using the at least one vector sequence sets as sequential input samples for training and execution of a pattern model that is learned to predicts for one or more future timesteps of interest, future relations for each node of the TKGs; (Jin [Page 2, Col. 2, Paragraph 4]: “Formally, we represent TKGs as sequences, and then build an autoregressive generative model on the sequences.” [Page 1, Col. 1, Paragraph 1]: “link prediction at future times … multi-step inference over future timestamps” Jin teaches that the sequences are used to build an autoregressive generative model, which uses link prediction at future times and multi-step inference over future timestamps, which is per node.)
training and execution of a forecasting model that is learned to predict, for one or more future timesteps of interests the nodes of the TKG associated with each of the predicted future relations; (Jin [Page 1, Col. 1, Paragraph 2]: “where each fact is represented as a triple of subject entity, object entity and the relation between them” [Page 1, Col. 1, Paragraph 1]: “Future facts can then be inferred in a sequential manner based on the two modules.” Jin teaches the triples consisting of subject entity, object entity, and the relation between them, and shows that future facts can be predicted based on this.)
using the predicted future relations and nodes to assemble predicted future TKGs describing a crime related scenario in an area of interest per future time steps of interest; (Jin [Page 7, Col. 2, Paragraph 2]: “RE-NET effectively infers new events using a powerful event encoder and an aggregator, and provides accurate prediction results” Jin teaches that RE-NET accurately infers future events, which is assembling a graph of future TKGs as they are structured as the aforementioned triples.)
…
…
Jin does not explicitly teach:
… a forecasting-based action recommendation system configured to iteratively compute one or more actions acting on the predicted future crime related scenario that steer the predicted future crime related scenario towards a desired scenario;
However, Jacobs teaches:
a forecasting-based action recommendation system configured to iteratively compute one or more actions acting on the predicted future crime related scenario that steer the predicted future crime related scenario towards a desired scenario; (Jacobs [Page 6, Col. 2, Paragraph 2]: “the ability to make process recommendations (e.g. events that should happen in the future to positively influence the outcome of cases)” Jacobs teaches the ability to make recommendations in the future to affect the scenario toward a desired one.)
Jin and Jacobs are in the same field of endeavor as the present invention, as the references are directed to making machine learning inferences using temporal knowledge graphs (TKGs) and other graph structured knowledge of events. It would have been obvious, before the effective filing date of the claimed invention, to a person of ordinary skill in the art, to combine forming event vectors consisting of triples in TKGs and making future predictions on events as taught in Jin with forecasting actions to steer toward a desired scenario as taught in Jacobs. Jacobs provides this additional functionality. As such, it would have been obvious to one of ordinary skill in the art to modify the teachings of Jin to include teachings of Jacobs because the combination would allow for the predictions of future events to be actionable. This has the potential benefit of enabling agents to act on the forecast of future events with the most optimal action towards a specific goal.
Jin and Jacobs do not explicitly teach:
…
one or more crime prediction processing devices that, alone or in combination, are configured to provide for execution of the following steps:
…
..
…
…
…
…
and control means configured to automatically adapt monitoring and/or surveillance devices deployed in the area of interest based on the computed one or more actions.
However, Muijs teaches:
…
one or more crime prediction processing devices that, alone or in combination, are configured to provide for execution of the following steps: (Muijs [¶ 0003]: “historical crime report including a set of historical crime data corresponding to a plurality of criminal events” [¶ 0012]: “incident database incorporates geographical data indicative of geographical regions of the historic incidents, and the weight factors are computed based on the geographical regions of the historic incidents”” Muijs teaches that historical crime data can be stored to predict incidents in the future.)
…
..
…
…
…
…
and control means configured to automatically adapt monitoring and/or surveillance devices deployed in the area of interest based on the computed one or more actions. (Muijs [¶ 0043]: “surveillance subunit which may further feed data related to past incidents to the database … camera-based surveillance subunit 230 may comprise a video processing circuit and may deploy intelligent edge- or server-based computer vision analytic” Muijs teaches a surveillance unit for control means and can capture video.)
Muijs is in the same field as the present invention, since it is directed to the prediction of incidents and crime using machine learning. It would have been obvious, before the effective filing date of the claimed invention, to a person of ordinary skill in the art, to combine the predictions of future events using TKGs as taught in Jin as modified by Jacobs with using these predictions to predict incidents/crime as taught in Muijs. Muijs provides this additional functionality. As such, it would have been obvious to one of ordinary skill in the art to modify the teachings of Jin as modified by Jacobs to include teachings of Muijs because the combination would allow for humans to better predict when/where a criminal event could occur. This has the potential benefit of improving the lives of many people, as caution could be taken to avoid such incidents/crime after it is forecasted.
Claim 2 is substantially similar to claim 1, but has the following additional limitations:
Regarding dependent claim 2, Jin, Jacobs, and Muijs teach:
A computer-implemented method for Temporal Knowledge Graph (TKG) forecasting, the method comprising: (Jin [Page 1, Col. 1, Paragraph 2]: “each fact may not be true forever and hence it is useful to associate each fact with a timestamp as a constraint, yielding a temporal knowledge graph (TKG)” Jin teaches that there is a collection/database of past timestamps associated with a fact, creating a TKG.)
… by a relation prediction module based on a graph database comprising scenario states in a number of past timesteps represented as TKGs, … (Jin [Page 1, Col. 1, Paragraph 2]: “where each fact is represented as a triple of subject entity, object entity and the relation between them” [Page 1, Col. 1, Paragraph 1]: “Future facts can then be inferred in a sequential manner based on the two modules.” Jin teaches that the TKG consists of triples consisting of subject entity, object entity, and the relation between them, and shows that future facts can be predicted based on this.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 3, Jin, Jacobs, and Muijs teach:
The method according to claim 2, further comprising: computing, by the entity prediction module, a time-dependent graph embedding for each relation contained in the TKGs; (Jin [Page 1, Col. 1, Paragraph 2]: “where each fact is represented as a triple of subject entity, object entity and the relation between them” [Page 1, Col. 1, Paragraph 1]: “Future facts can then be inferred in a sequential manner based on the two modules.” Jin teaches that the TKG consists of triples consisting of subject entity, object entity, and the relation between them, and shows that future facts can be predicted based on this.)
and computing, by the entity prediction module based on a predefined similarity metric, a similarity of each relation contained in the TKGs to each of the other relations contained in the TKGs and creating, based on the computed similarities, relation similarity matrices for each relation and for each timestep. (Jacobs [Page 4, Col. 1, Paragraph 4]: “KBlrn embeds the graph structure into the latent space which allows us to obtain a fixed-vector representation of each node as needed by the ProcK framework” Jacobs teaches that the graph structure is embedded into a latent space, which is where similar items can be grouped closer together. The set of all relations of similar triples is analogous to a relation similarity matrix for each relation and for each timestep.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 4, Jin, Jacobs, and Muijs teach:
The method according to claim 3, wherein the graph embedding for the relationships is computed on the graph database including scenario states of all available past timesteps. (Jacobs [Page 3, Col. 2, Paragraph 4]: “The input to the graph embedder is the graph G = (V, R, E)” Jacobs teaches that the input to the embedder is the entire graph, showing that the full graph database including all past timesteps is embedded.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 5, Jin, Jacobs, and Muijs teach:
The method according to claim 3, further comprising: using the relation similarity matrices as sequential input samples for training and execution of the forecasting model that is learned to predict, based on past similarity matrices for a relation, future similarity matrices for the respective relation. (Jin [Page 1, Col. 1, Paragraph 2]: “where each fact is represented as a triple of subject entity, object entity and the relation between them” [Page 1, Col. 1, Paragraph 1]: “Future facts can then be inferred in a sequential manner based on the two modules.” Jin teaches the triples consisting of subject entity, object entity, and the relation between them, and shows that future facts can be predicted based on this.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 6, Jin, Jacobs, and Muijs teach:
The method according to claim 2, wherein the forecasting model is a sequential model implemented in form of a recurrent neural network (RNN). (Jin [Page 3, Col. 1, Paragraph 1]: “RE-NET consists of a Recurrent Neural Network (RNN) as a recurrent event encoding module and a neighborhood aggregation module to capture the information of graph structures” Jin teaches that the RE-NET model is a RNN.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 7, Jin, Jacobs, and Muijs teach:
The method according to claim 5, further comprising: extracting, from the predicted future similarity matrices, for each predicted future relation an entity with a highest predicted similarity and associating the extracted entity with the respective [subject, relation] or [relation, object] pair to obtain all predicted triples of future TKGs for timesteps of interest. (Jin [Page 5, Algorithm 1]: Jin teaches in step 4 of the algorithm that the triples are ranked and the top-k are chosen. The criteria, as mentioned above that the embeddings are in latent space, has a similarity component to them.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 8, Jin, Jacobs, and Muijs teach:
The method according to claim 7, wherein the extraction of a predicted entity for a predicted future relation takes into consideration the neighborhood of the respective entity across all timesteps by means of a neighborhood aggregation matrix created by computing, and subsequently aggregating over, the similarity of each relation in the neighborhood of the predicted entity across all timesteps to the respective predicted future relation. (Jin [Page 1, Col. 1, Paragraph 1]: “RE-NET employs a recurrent event encoder to encode past facts, and uses a neighborhood aggregator to model the connection of facts at the same timestamp” [Page 3, Col. 2, Paragraph 3]: “where e_{s}, e_{r} in R^{d} are learnable embedding vectors specified for subject entity s and relation r.” Jin teaches that the RE-NET encodes connections of facts at a given timestamp and are vectors detailing the relation. This is done across the neighborhood of entity across timestamps.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 9, Jin, Jacobs, and Muijs teach:
The method according to claim 2, further comprising: encoding the generated relationship vectors for dimensionality reduction prior to creating the sequential dataset, and/or encoding the relation similarity matrices for dimensionality reduction prior to providing the relation similarity matrices to the forecasting model. (Jin [Page 11, Col. 2, Paragraph 2]: “we adopt the block-diagonal decomposition … where each relation-specific weight matrix is decomposed into a block-diagonal by decomposing into low-dimensional matrices.” Jin teaches a decomposition of the matrices such that the relation matrix with the vectors is decomposed into lower dimensions.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 10, Jin, Jacobs, and Muijs teach:
The method according to claim 2, further comprising: embedding the relation prediction module and the entity prediction module in a forecasting-based action recommendation system that iteratively computes appropriate actions acting on a predicted scenario that steer the predicted scenario towards a desired scenario. (Jacobs [Page 6, Col. 2, Paragraph 2]: “the ability to make process recommendations (e.g. events that should happen in the future to positively influence the outcome of cases)” Jacobs teaches the ability to make recommendations in the future to affect the scenario toward a desired one.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 11, Jin, Jacobs, and Muijs teach:
The method according to claim 2, wherein the graph database is a database containing crime related scenario states in a number of past timesteps represented as TKGs, the method further comprising: predicting future TKGs describing a crime related scenario in an area of interest per future time steps of interest, (Muijs [¶ 0003]: “historical crime report including a set of historical crime data corresponding to a plurality of criminal events” [¶ 0012]: “incident database incorporates geographical data indicative of geographical regions of the historic incidents, and the weight factors are computed based on the geographical regions of the historic incidents”” Muijs teaches that historical crime data can be stored to predict incidents in the future.)
iteratively computing one or more actions acting on the predicted future crime related scenario that steer the predicted future crime related scenario towards a desired scenario, (Jacobs [Page 6, Col. 2, Paragraph 2]: “the ability to make process recommendations (e.g. events that should happen in the future to positively influence the outcome of cases)” Jacobs teaches the ability to make recommendations in the future to affect the scenario toward a desired one.)
and automatically adapting monitoring and/or surveillance devices deployed in the area of interest based on the computed one or more actions. (Muijs [¶ 0043]: “surveillance subunit which may further feed data related to past incidents to the database … camera-based surveillance subunit 230 may comprise a video processing circuit and may deploy intelligent edge- or server-based computer vision analytic” Muijs teaches a surveillance unit for control means and can capture video.)
The reasons to combine are substantially similar to those of claim 1.
Regarding dependent claim 12, Jin, Jacobs, and Muijs teach:
The method according to claim 2, wherein the graph database is a database of a public health related infection control service containing infection related scenario states in a number of past timesteps represented as TKGs, the method further comprising: predicting future TKGs describing infection development on a predefined geographical level per future time steps of interest, (Muijs [¶ 0007]: “enable a more effective predicting of an incident in that region” The field of use is changed in this limitation to public health, which may fall under an “incident” as taught by Muijs.)
iteratively computing one or more actions acting on the predicted future infection related scenario that steer the predicted future scenario towards a desired scenario, (Jacobs [Page 6, Col. 2, Paragraph 2]: “the ability to make process recommendations (e.g. events that should happen in the future to positively influence the outcome of cases)” Jacobs teaches the ability to make recommendations in the future to affect the scenario toward a desired one.)
and automatically adapting access control devices for public buildings, advertising on digital advertising panels, and/or frequency of public transport based on the computed one or more actions. (Muijs [¶ 0043]: “surveillance subunit which may further feed data related to past incidents to the database … camera-based surveillance subunit 230 may comprise a video processing circuit and may deploy intelligent edge- or server-based computer vision analytic” Muijs teaches a surveillance unit for control means and can capture video.)
The reasons to combine are substantially similar to those of claim 1.
Claim 13 is substantially similar to claim 1, but has the following additional limitations:
Regarding independent claim 13, Jin, Jacobs, and Muijs teach:
A processing system comprising one or more processors which, alone or in combination, are configured to provide for execution of a method for Temporal Knowledge Graph; TKG) forecasting, the method comprising: (Muijs [¶ 0024]: “a computer program product is provided comprising instructions for causing a processor system to perform the method.” Muijs teaches a processor that can use instructions.)
… by a relation prediction module based on a graph database comprising scenario states in a number of past timesteps represented as TKGs, … (Jin [Page 1, Col. 1, Paragraph 2]: “each fact may not be true forever and hence it is useful to associate each fact with a timestamp as a constraint, yielding a temporal knowledge graph (TKG)” Jin teaches that there is a collection/database of past timestamps associated with a fact, creating a TKG.)
The reasons to combine are substantially similar to those of claim 1.
Claim 14 is rejected on the same grounds under 35 U.S.C. 103 as claim 1 as they are substantially similar. Mutatis mutandis.
Claim 15 is substantially similar to claim 1, but has the following additional limitations:
Regarding independent claim 15, Jin, Jacobs, and Muijs teach:
A tangible, non-transitory computer-readable medium having instructions thereon which, upon being executed by one or more processors, alone or in combination, provide for execution of a method of Temporal Knowledge Graph (TKG) forecasting, the method comprising: (Muijs [¶ 0024]: “a computer program product is provided comprising instructions for causing a processor system to perform the method.” Muijs teaches a processor that can use instructions.)
… by a relation prediction module based on a graph databased comprising scenario states in a number of past timesteps represented as TKGs, … (Jin [Page 1, Col. 1, Paragraph 2]: “each fact may not be true forever and hence it is useful to associate each fact with a timestamp as a constraint, yielding a temporal knowledge graph (TKG)” Jin teaches that there is a collection/database of past timestamps associated with a fact, creating a TKG.)
… by an entity prediction module, … (Jin [Page 2, Col. 2, Paragraph 4]: “Formally, we represent TKGs as sequences, and then build an autoregressive generative model on the sequences.” [Page 1, Col. 1, Paragraph 1]: “link prediction at future times … multi-step inference over future timestamps” Jin teaches that the sequences are used to build an autoregressive generative model, which uses link prediction at future times and multi-step inference over future timestamps, which is per node.)
The reasons to combine are substantially similar to those of claim 1.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KYU HYUNG HAN whose telephone number is (703) 756-5529. The examiner can normally be reached on MF 9-5.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Alexey Shmatov can be reached on (571) 270-3428. 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.
/Kyu Hyung Han/
Examiner
Art Unit 2123
/ALEXEY SHMATOV/Supervisory Patent Examiner, Art Unit 2123
Prosecution Insights