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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Beginning the analysis with claim 12, the claim recites a method, which is a statutory category of invention (Step 1: YES).
the claim recites a method of interpreting a multi-dimensional dataset (interpreting data encompasses mental activity practically performed in the human mind using observation, evaluation, judgement, and opinion). The method includes steps of:
determining a correlation of the one or more dimensions of the multi-dimensional dataset to the results output by each of the plurality of machine learning models (determining a correlation encompasses mental activity practically performed in the human mind using observation, evaluation, judgement, and opinion; For example, a human observing the output of the models could mentally determine which dimensions were correlated);
selecting, based on the correlation determined between the one or more dimensions and the result output by each of the plurality of machine learning models, a subset of the plurality of machine learning models to obtain the result for each of the subset of the plurality of machine learning models (selecting a subset of the machine learning models encompasses mental activity practically performed in the human mind using observation, evaluation, judgement, and opinion; For example, a human could mentally select a set of machine learning models by observing the correlations); and
outputting the result for each of the subset of the plurality of machine learning models (outputting the result, without further limitation, covers any expression of the information provided by the machine learning models; Therefore, this step encompasses mental activity practically performed in the human mind using observation, evaluation, judgement, and opinion; For example, a human could speak or write the observed results of the machine learning models).
As shown above, claim 12 recites a series of steps that could practically be performed in the human mind. Claim 12 therefore recites an abstract idea (Step 2A, Prong One: YES)
This judicial exception is not integrated into a practical application because the only additional elements beyond the abstract idea itself are the recited “machine learning models”. Claim 12 recites a step of applying a plurality of machine learning models to the multi-dimensional dataset to obtain a result output by each of the plurality of machine learning models. However, merely applying generic machine learning models to a dataset to generate generic output amounts to mere instructions to apply an exception. The claim recites the idea of a solution or outcome (“output”) without reciting details of how the machine learning models generate that output. See MPEP 2106.05(f). Once the machine learning models have been “applied” to the multi-dimensional dataset, the remainder of the claim encompasses purely mental steps of determining a correlation, selecting, and outputting a result based on the observed output of the machine learning models. Therefore, even when considering the claim as a whole, the generic “machine learning models” applied to a multi-dimensional dataset to generate generic “output” do not integrate the recited judicial exception into a practical application (Step 2A, Prong Two: NO), and the claim is directed to the judicial exception (Step 2A: YES).
The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because, as described above, the recited machine learning models are recited at a high level of generality and amount to no more than mere instructions to apply the judicial exception using generic machine learning models. Even when considered in combination, the claim as a whole does not provide an inventive concept (Step 2B: NO).
Independent claim 1 is directed to a “device” comprising a memory and one or more processors to store the multi-dimensional dataset and to perform the method of claim 12. Independent claim 20 is directed to a non-transitory computer-readable storage medium comprising instructions that, when executed, cause one or more processors to perform the method of claim 12. As discussed in MPEP 2106.04(a)(2), claims can recite an mental process even if they are claimed as being performed in a computer. Claim 1 merely performs the mental process of claim 12 using a generic computer. Similarly, both product claims and process claims may recite mental processes. The analysis applied above to claim 12 therefore applies equally to both claims 1 and 20, and they are rejected for the same reasons as claim 12.
With respect to claims 2 and 13, outputting the result in plain language encompasses mental activity practically performed in the human mind using observation, evaluation, judgement, and opinion. For example, a human could speak or write the observed results of the machine learning models in English. The claims do not recite any additional elements and are therefore directed to an abstract idea without significantly more for the same reasons as claims 1 and 12.
With respect to claims 3-4 and 14-15, outputting the result as an impact graph encompasses mental activity practically performed in the human mind using observation, evaluation, judgement, and opinion. For example, a human could or write the observed results of the machine learning models as an impact graph. The claims do not recite any additional elements and are therefore directed to an abstract idea without significantly more for the same reasons as claims 1 and 12.
With respect to claims 5 and 16, outputting the result as a graphical representation of a decision tree encompasses mental activity practically performed in the human mind using observation, evaluation, judgement, and opinion. For example, a human could or write the observed results of the machine learning models by drawing a decision tree. The claims do not recite any additional elements and are therefore directed to an abstract idea without significantly more for the same reasons as claims 1 and 12.
With respect to claims 6-7 and 17-18, the claims require comparing the correlations to determine low relevance dimensions and outputting an indication explaining that the low relevance dimensions have low relevance to the result as a sentence in plain language. Comparing the correlations encompasses mental activity practically performed in the human mind using observation, evaluation, judgement, and opinion. For example, a human could mentally compare the correlations and mentally determine one or more associated dimensions with the lowest correlations. Additionally, outputting an explanation as a sentence in plain language could be performed by a human speaking or writing the explanation. The claims do not recite any additional elements and are therefore directed to an abstract idea without significantly more for the same reasons as claims 1 and 12.
Claims 8 and 19 require refraining from transforming the one or more dimensions of the multi-dimensional dataset prior to application of the plurality of machine learning models. Refraining from performing any action encompasses mental activity performed in the human mind, because it would merely require a human to not act (in this case, to not perform a transformation). The claims do not recite any additional elements and are therefore directed to an abstract idea without significantly more for the same reasons as claims 1 and 12.
With respect to claims 9 and 10, the claims require determining one or more of a plurality of charts to explain the corresponding result, ranking the charts, and selecting and outputting the highest ranked chart along with an explanation in plain language. Each of these steps encompasses mental activity practically performed in the human mind using observation, evaluation, judgement, and opinion. For example, a human could manually draw a plurality of charts, mentally rank them, and select the chart mentally determined to have the highest rank, then provide a spoken or written explanation describing the chart. The claims do not recite any additional elements and are therefore directed to an abstract idea without significantly more for the same reasons as claim 1.
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.
Claim(s) 1-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Patten, Jr. et al. (U.S. Patent Application Pub. No. 2019/0362417, hereinafter “Patten”), in view of Rope et al. (U.S. Patent No. 10,395,215, hereinafter “Rope”).
In regard to claim 1, Patten discloses a device configured to interpret a multi-dimensional dataset (Fig. 3, 300), the device comprising:
a memory configured to store the multi-dimensional dataset (hard drive 311, ROM 310, etc., paragraph [0066]); and]
one or more processors (processing unit 303, paragraph [0065]) configured to:
apply a plurality of machine learning models to the multi-dimensional dataset to obtain a result output by each of the plurality of machine learning models (Fig. 1, a dataset 104 comprising multiple dimensions (account numbers, transaction ID, transaction amount, etc.) is analyzed by a plurality of analytical models to produce analytic results, paragraphs [0041-0042]; the analytical models comprising machine learning models, paragraphs [0061-0063]);
determine a correlation of one or more dimensions of the multi-dimensional dataset to the results output by each of the plurality of machine learning models (the analytical results are further analyzed to determine correlation relationships, global variable importance, etc., paragraph [0054]);
output the result for each of the plurality of machine learning models (the analysis results are provided to a natural language generator and an output report is generated, paragraph [0049]).
While Patten discloses “key elements” of the analysis are output (paragraph [0049]), Patten does not expressly disclose selecting a subset of the plurality of machine learning models.
Rope discloses a device configured to interpret a multi-dimensional dataset, programmed to:
select, based on the correlation determined between the one or more dimensions and the result output by each of the plurality of machine learning models, a subset of the plurality of machine learning models to obtain the result for each of the subset of the plurality of machine learning models (statistical computation constructs an interestingness index the results of predictive models, column 3, lines 31-45; the statistical computations including correlation, column 4, lines 15-23; where key findings are determined by selecting a subset of results according to the interestingness index, column 7, lines 1-12).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to select a subset of the results output by the machine learning models based on the correlation determined between the one or more dimensions and the result output, because it would allow business users to easily understand the most important aspects of the analysis, as suggested by Rope (column 2, lines 49-57).
In regard to claim 2, Patten discloses the one or more processors are configured to output the result as a sentence using plain language (natural language, paragraph [0049]).
In regard to claim 3, Patten discloses the one or more processors are configured to output the result for at least one of the subset of the plurality of machine learning models as a graph identifying a relevance of each of the one or more dimensions to the result for each of the subset of the plurality of machine learning models (visualizations comprising relevant graphs, paragraph [0054]).
In regard to claim 4, Patten discloses the graph comprises an impact graph (graph of global variable importance, paragraph [0054]).
In regard to claim 5, Patten discloses the one or more processors are configured to output the result for each of the subset of the plurality of machine learning models as a graphical representation of a decision tree (decision tree visual representation, paragraph [0054]).
In regard to claim 6, Patten does not disclose determining one or more low relevance dimensions of the multi-dimensional dataset.
Rope discloses to:
determine, based on a comparison of the correlation determined between the one or more dimensions and the result output by each of the plurality of machine learning models to a relevance threshold, one or more low relevance dimensions of the multi-dimensional dataset that have low relevance to the result output by each of the plurality of machine learning models (based on an overly high correlation, fields of the dataset that are redundant are identified, column 10, line 62 to column 11, line 4); and
output an indication explaining that the one or more low relevance dimensions have low relevance to the result (an output explaining that the fields are redundant, column 11, lines 1-4).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to determine one or more low relevance dimensions, because it would allow business users to easily understand the most important aspects of the analysis, as suggested by Rope (column 2, lines 49-57).
In regard to claim 7, Patten does not disclose explaining the one or more low relevance dimensions having low relevance to the result.
Rope discloses to output a sentence in plain language that explain the one or more low relevance dimensions having low relevance to the result (a natural language output explaining that the fields are redundant, column 11, lines 1-4).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to output a sentence in plain language that explain the one or more low relevance dimensions having low relevance to the result, because it would allow business users to easily understand the most important aspects of the analysis, as suggested by Rope (column 2, lines 49-57).
In regard to claim 8, Patten discloses the one or more processors are further configured to refrain from transforming the one or more dimensions of the multi-dimensional dataset prior to application of the plurality of machine learning models (Patten discloses “in some embodiments” visualizations will be transformed variants of the applicable model, paragraph [0054]; therefore, in other embodiments, the processers must refrain from performing these transforms).
In regard to claim 9, Patten discloses the one or more processors are further configured to:
determine, based on the results for each of the one or more of the plurality of machine learning models, one or more of a plurality of charts to explain the corresponding result (relevant visualizations associated with the models are generated, paragraph [0054]).
While Patten discloses to select relevant visualizations for output (paragraph [0055]), Patten does not expressly disclose ranking the one or more of the plurality of charts.
Rope discloses to:
rank the one or more of the plurality of charts to identify a highest ranked chart; select the highest ranked chart (from a large number of generated tables, graphs, etc., the top ranked graphics are selected for output, column 9, lines 55-64 and column 10, lines 20-32); and
output the highest ranked chart as a visual chart (see Fig. 12, visualization 1210, column 11, lines 13-19).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to rank the one or more of the plurality of charts to identify a highest ranked chart and select and output the highest rank chart, because it would allow business users to easily understand the most important aspects of the analysis, as suggested by Rope (column 2, lines 49-57).
In regard to claim 10, Patten does not disclose to generate an explanation in plain language explaining a formulation of the visual chart.
Rope discloses to:
generate an explanation in plain language explaining a formulation of the visual chart (Fig. 12, text insight 1212 of visualization 1210, column 11, lines 13-19); and
output the explanation (see Fig. 12).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to generate and output an explanation in plain language explaining a formulation of the visual chart, because it would allow business users to easily understand the most important aspects of the analysis, as suggested by Rope (column 2, lines 49-57).
In regard to claim 11, Patten discloses the one or more processors are further configured to:
generate a pipeline report explaining how the device produced the plurality of the machine learning models (analytical results generating information about how the results were achieved, paragraph [0039]); and
output the pipeline report (output report, paragraph [0039]).
In regard to claim 12, Patten discloses a method of interpreting a multi-dimensional dataset (Fig. 3, 300), the method comprising:
applying a plurality of machine learning models to the multi-dimensional dataset to obtain a result output by each of the plurality of machine learning models (Fig. 1, a dataset 104 comprising multiple dimensions (account numbers, transaction ID, transaction amount, etc.) is analyzed by a plurality of analytical models to produce analytic results, paragraphs [0041-0042]; the analytical models comprising machine learning models, paragraphs [0061-0063]);
determining a correlation of one or more dimensions of the multi-dimensional dataset to the results output by each of the plurality of machine learning models (the analytical results are further analyzed to determine correlation relationships, global variable importance, etc., paragraph [0054]);
outputting the result for each of the plurality of machine learning models (the analysis results are provided to a natural language generator and an output report is generated, paragraph [0049]).
While Patten discloses “key elements” of the analysis are output (paragraph [0049]), Patten does not expressly disclose selecting a subset of the plurality of machine learning models.
Rope discloses a method of interpreting a multi-dimensional dataset, comprising:
selecting, based on the correlation determined between the one or more dimensions and the result output by each of the plurality of machine learning models, a subset of the plurality of machine learning models to obtain the result for each of the subset of the plurality of machine learning models (statistical computation constructs an interestingness index the results of predictive models, column 3, lines 31-45; the statistical computations including correlation, column 4, lines 15-23; where key findings are determined by selecting a subset of results according to the interestingness index, column 7, lines 1-12).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to select a subset of the results output by the machine learning models based on the correlation determined between the one or more dimensions and the result output, because it would allow business users to easily understand the most important aspects of the analysis, as suggested by Rope (column 2, lines 49-57).
In regard to claim 13, Patten discloses outputting the result comprises outputting the result as a sentence using plain language (natural language, paragraph [0049]).
In regard to claim 14, Patten discloses outputting the result comprises outputting the result for at least one of the subset of the plurality of machine learning models as a graph identifying a relevance of each of the one or more dimensions to the result for each of the subset of the plurality of machine learning models (visualizations comprising relevant graphs, paragraph [0054]).
In regard to claim 15, Patten discloses the graph comprises an impact graph (graph of global variable importance, paragraph [0054]).
In regard to claim 16, Patten discloses outputting the result comprises outputting the result for each of the subset of the plurality of machine learning models as a graphical representation of a decision tree (decision tree visual representation, paragraph [0054]).
In regard to claim 17, Patten does not disclose determining one or more low relevance dimensions of the multi-dimensional dataset.
Rope discloses determining, based on a comparison of the correlation determined between the one or more dimensions and the result output by each of the plurality of machine learning models to a relevance threshold, one or more low relevance dimensions of the multi-dimensional dataset that have low relevance to the result output by each of the plurality of machine learning models (based on an overly high correlation, fields of the dataset that are redundant are identified, column 10, line 62 to column 11, line 4); and
outputting an indication explaining that the one or more low relevance dimensions have low relevance to the result (an output explaining that the fields are redundant, column 11, lines 1-4).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to determine one or more low relevance dimensions, because it would allow business users to easily understand the most important aspects of the analysis, as suggested by Rope (column 2, lines 49-57).
In regard to claim 18, Patten does not disclose explaining the one or more low relevance dimensions having low relevance to the result.
Rope discloses outputting a sentence in plain language that explain the one or more low relevance dimensions having low relevance to the result (a natural language output explaining that the fields are redundant, column 11, lines 1-4).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to output a sentence in plain language that explain the one or more low relevance dimensions having low relevance to the result, because it would allow business users to easily understand the most important aspects of the analysis, as suggested by Rope (column 2, lines 49-57).
In regard to claim 19, Patten discloses refraining from transforming the one or more dimensions of the multi-dimensional dataset prior to application of the plurality of machine learning models (Patten discloses “in some embodiments” visualizations will be transformed variants of the applicable model, paragraph [0054]; therefore, in other embodiments, the processers must refrain from performing these transforms).
In regard to claim 20, Patten discloses a non-transitory computer-readable storage medium storing instructions (paragraph [0015]) that, when executed, cause one or more processors to:
apply a plurality of machine learning models to the multi-dimensional dataset to obtain a result output by each of the plurality of machine learning models (Fig. 1, a dataset 104 comprising multiple dimensions (account numbers, transaction ID, transaction amount, etc.) is analyzed by a plurality of analytical models to produce analytic results, paragraphs [0041-0042]; the analytical models comprising machine learning models, paragraphs [0061-0063]);
determine a correlation of one or more dimensions of the multi-dimensional dataset to the results output by each of the plurality of machine learning models (the analytical results are further analyzed to determine correlation relationships, global variable importance, etc., paragraph [0054]);
output the result for each of the plurality of machine learning models (the analysis results are provided to a natural language generator and an output report is generated, paragraph [0049]).
While Patten discloses “key elements” of the analysis are output (paragraph [0049]), Patten does not expressly disclose selecting a subset of the plurality of machine learning models.
Rope discloses a device configured to interpret a multi-dimensional dataset, programmed to:
select, based on the correlation determined between the one or more dimensions and the result output by each of the plurality of machine learning models, a subset of the plurality of machine learning models to obtain the result for each of the subset of the plurality of machine learning models (statistical computation constructs an interestingness index the results of predictive models, column 3, lines 31-45; the statistical computations including correlation, column 4, lines 15-23; where key findings are determined by selecting a subset of results according to the interestingness index, column 7, lines 1-12).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to select a subset of the results output by the machine learning models based on the correlation determined between the one or more dimensions and the result output, because it would allow business users to easily understand the most important aspects of the analysis, as suggested by Rope (column 2, lines 49-57).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Achin et al., Baran et al., Birnbaum et al., Hugot et al., Iyer et al., Kass et al., and Lyras disclose additional methods for explaining datasets in plain language.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRIAN LOUIS ALBERTALLI whose telephone number is (571)272-7616. The examiner can normally be reached M-F 8AM-3PM, 4PM-5PM.
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, Bhavesh Mehta can be reached at 571-272-7453. 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.
BLA 3/31/26
/BRIAN L ALBERTALLI/ Primary Examiner, Art Unit 2656