DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Amendments
This office action responds to the amendments filed on July 22, 2025 for application 18/278,874. Claims 1-4, 6-10, and 12-14 are amended, claim 5 is cancelled, and claims 1-4 and 6-15 remain pending in the application.
Response to Arguments
The Examiner has fully considered the Applicant’s arguments filed on July 22, 2025, and the Examiner responds as provided below.
Regarding the Applicant’s response at page 6 of the Remarks that concerns the objection to the claims, the claim amendments are sufficient to overcome the objection except in one case. Claim 11 was not amended, and the objection remains as detailed below for claim 11.
Regarding the Applicant’s response at page 6 of the Remarks that concerns the § 112 rejection to claims 3 and 4, the amendments to the claims are sufficient to address the issues of indefiniteness. Thus, the § 112(b) rejection is withdrawn.
Regarding the Applicant’s response at pages 6-8 of the Remarks that concerns the § 101 rejection, the Examiner respectfully submits that the arguments are not persuasive and the § 101 rejection is maintained.
The Applicant initially argues that “Applicant’s invention sets forth an improved technique for increasing privacy” and “the claimed invention improves another technology.” Remarks at 6-7 (citing MPEP § 2106.04(d)(1)) (emphasis retained). Although Applicant correctly notes that a claimed invention possessing an abstract idea is patentable if it represents a technological advancement, Applicant’s claimed invention fails to advance any perceptible technology. Independent claims 1, 14, and 15 broadly claim mathematical relationships, but they fail to recite any sufficient computer or network environment—outside of the recitation of generic computer elements at a high level--that would consequently amount to a practical application under Step 2A, Prong II. Applicant’s claims fail to “impose[] meaningful limit[s] on the judicial exception” of mathematical expressions, and the claims amount to “a drafting effort designed to monopolize the judicial exception.” See MPEP § 2106.04(d).
Applicant states, “Applicant’s claimed invention takes an input dataset from which private information could still be leaked and outputs a private dataset, which has practically relevant and reliable privacy and user protection properties.” Remarks at 7. These two steps, which respectively amount to the gathering of data and the output of data, are pre-solution and post-solution activity that fail to transform the judicial exception of mathematical expressions into patentable subject matter. See MPEP § 2106.05(g). Regarding the argument pertaining to the “determining” steps, these steps essentially represent mathematical relationships that are ineligible subject matter even though they can be employed to protect data. Again, the independent claims lack specific limitations that transform the mathematical expressions into a practical application, and thus an identifiable technological improvement within the claimed subject matter.
Accordingly, the § 101 rejection is maintained as detailed below.
Regarding the Applicant’s response at pages 8-9 of the Remarks that concerns the § 103 rejection, the Applicant bases their argument on the Karhunen-Love transform. None of the claims, including independent claims 1, 14, and 15, recite the Karhunen-Love transform. Instead, the claims merely recite the broad limitation “the transform depends on the plurality of data points,” and the Applicants have inappropriately imported a limitation from the specification into the claims. See MPEP § 2111.01(II) (stating “It is improper to import claim limitations from the specification”). Given the broad nature of the contested limitation, the use of Lopez to teach this broad limitation is appropriate. If the Applicant elects to recite the Karhunen-Love transform, then Applicant’s argument with respect to Lopez will have merit.
Accordingly, the § 103 rejection is maintained as detailed below.
Claim Objections
Claim 11 is objected to because of the following informalities: each claim should begin with “The method…” Appropriate correction is required.
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.
Regarding Independent Claims 1, 14, and 15
Independent claims 1, 14, and 15 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claims 1, 14, and 15 recite a process that encompasses a mathematical concept. See MPEP § 2106.04(a)(2)(I) (stating under Step 2A, Prong I that “[t]he mathematical concepts grouping is defined as mathematical relationships, mathematical formulas or equations, and mathematical calculations.”). The claims recite steps that are primarily used to implement differential privacy. Although the claims do not incorporate a literal mathematical expression, “a mathematical concept [within a claim] need not be expressed in mathematical symbols….” Id. Here, a mathematical concept exists because a dataset is being manipulated through the mathematical application of a transform. Thus, although the claims do not literally possess a “mathematical formula[] or equation[],” the claims incorporate the well-known use of a “mathematical relationship” via the use of a transform. See MPEP § 2106.04(a)(2)(I)(A) (stating “A mathematical relationship is a relationship between variables or numbers. A mathematical relationship may be expressed in words or using mathematical symbols.”).
This judicial exception is not integrated into a practical application because under Step 2A, Prong II because claims 1, 14, and 15 fail to yield claimed subject matter that encompasses an “improvement in the functioning of a computer, or an improvement to other technology or technical field.” MPEP § 2106.04(d)(I). The claims recite several steps taken to generically prevent privacy leakage between “data processing devices,” but the claims do not amount to improving the functioning of a computer system or improving a technology. Although the claimed steps associated with determining a privacy leakage may serve as the basis for an improvement of a computer system, the claims as currently presented fail to yield a claimed improvement to a computer system or technology. More specifically, the independent claims merely recite “data processing devices,” within the preamble of the claims, and this recital of computer elements at a high level of generality1 fails to yield an improvement to the “improvement in the functioning of a computer, or an improvement to other technology or technical field.”.
Finally, claims 1, 14, and 15 do not include additional elements under step 2B that are sufficient to amount to significantly more than the judicial exception. Under step 2B, an inventive concept can be found when the claimed subject matter yields an improvement to the functioning of a computer. See MPEP § 2106.05(I)(A) (stating “Limitations that the courts have found to qualify as "significantly more" when recited in a claim with a judicial exception include: i. Improvements to the functioning of a computer,...”). Because the subject matter of the claims is intended to be applied to the prevention of privacy leakage, the presence of an inventive concept in the claimed subject matter is a substantial possibility. However, the claims as currently presented fail to recite elements to establish an inventive concept involving a specific circumstance, such as a multiparty communication or a querying of a database. Instead, the claims merely recite a mathematical framework that can be applied to any computer system. See id. (stating “Adding the words ‘apply it’ (or an equivalent) with the judicial exception, or mere instructions to implement an abstract idea on a computer,...”).
Regarding Dependent Claims 2-13
The dependent claims recite the following mathematical limitations:
Claim 2: decorrelation efficiency metric value, maximation of transform coefficients
Claim 3: complexity threshold
Claim 4: utility metric value, utility threshold
Claim 6: orthogonal transform
Claim 7: variance value, probabilistic model
Claim 8: total sum, transform coefficients, variance threshold
Claim 9: lowest non-zero variance value, low-variance transform coefficient
Claim 10: private transform coefficients, independent noise
Claim 11: Laplace distribution noise
Claim 12: noise distribution parameter
Claim 13: datapoints, inter-user correlations
Dependent claims 2-13 merely extend the claiming of the mathematical concept captured in claim 1. In other words, dependent claims 2-13 fail to recite any limitations associated with the claiming of a computer system that yield a practical application or an inventive concept under steps 2A and 2B, respectively. Accordingly, dependent claims 2-13 are rejected under the same rationale as previously provided for claims 1, 14, and 15.
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.
The following conventions apply to the mapping of the prior art to the claims:
Italicized text – claim language.
Parenthetical plain text – Examiner’s citation and explanation.
Citation without an explanation – an explanation has been previously provided for the respective limitation(s).
Quotation marks – language quoted from a prior art reference.
Underlining – language quoted from a claim.
Brackets – material altered from either a prior art reference or a claim, which includes the Examiner’s explanation that relates a claim limitation to the quoted material of a reference.
Braces – a limitation taught by another reference, but the limitation is presented with the mapping of the instant reference for context.
Numbered superscript – a first phrase to be moved upwards to the primary reference analysis.
Lettered superscript – a second phrase to be moved after the movement of the first phrase from which it was lifted, or more succinctly, move numbered material first, lettered material last.
A. Claims 1 and 6-15 are rejected under 35 U.S.C. 103 as being unpatentable over Amroabadi (US 2021/0133590, “Amroabadi”) in view of Peterson et al. (US 2021/0073677, “Peterson”), and further in view of Lopez (US 2017/0104610, “Lopez”).
Regarding Claim 1
Amroabadi discloses
A method for increasing privacy of user data of a plurality of users within a dataset (¶¶ [0046]-[0048], “Differential privacy (also referred to as “DP” herein) is a property of a data releasing mechanism [method] which requires that presence of a particular individual's [user’s] data in a dataset [of a plurality of users] must have a limited effect on the output of any output resulting from the analysis of the data. Differential privacy is usually achieved by injecting calibrated noise [that thereby increases privacy] to the releasing mechanism outputs.”),
the method (¶¶ [0046]-[0048]) comprising, in one or more data processing devices (¶ [0210], “System 100, in particular, one or more [plurality] of autoencoder 102 and privatizer 104, may be implemented as software and/or hardware, for example, in a computing [data processing] device 120 as illustrated in FIG. 14.”):
providing a dataset (¶ [0129], “Input data 110, illustrated as ‘Original Data’ in FIG. 1, can include data to be privatized such as time-series data [dataset] represented as a series query.”) comprising a plurality of data points of a plurality of users and comprising inter-user correlations within the plurality of data points (¶ [0022], “Time series stamping of individual client data is a unique characteristic that makes existing (cross-sectional) privacy [of a plurality of users] mechanisms inadequate, since successive timestamps [of data points involving different users, i.e., inter-users] from the same source can be highly correlated. This correlation can cause answers to different queries over time series data to also become correlated, e.g., a sequence of queries computing the average weight of a community at successive weeks.”);
determining a plurality of transform coefficients by applying a transform on the plurality of data points (¶ [0085], “To remedy this limitation, and answer multiple queries over time-series data under differential privacy, [16] introduces the FPAk algorithm that perturbs the Discrete Fourier Transform (DFT) of the query answers. For answering n queries, FPAk improves the error from θ(n) (error of standard differential privacy techniques) to roughly θ(k) where k is the number of DFT coefficients that can (approximately) reconstruct all the n query answers.”; and Fig. 1, ¶¶ [0139]-[0140], “Encoder 112 can be configured to present input data [datapoints] 110 in a sparse way, for example, represented with few large coefficients or elements, and lots of zeroes, with a sparse domain.”),
2 …;
determining a plurality of private transform coefficients from the plurality of transform coefficients… 1 (Fig. 1, ¶ [0146], “This robustness may increase the utility of private data 140 (private signals) generated by decoder 113 where Laplace noise 114 is added to the latent [private transform] coefficients during the privatization mechanism of privatizer 104 [employing differential privacy as disclosed by Peterson].”)
to each non-zero transform coefficient of the plurality of transform coefficients (¶ [0154], “In some embodiments, the added noise follows a Laplacian distribution, such as a Lap (k1/2Δ2(F)/ϵ) distribution, where k is the number of non-zero coefficients used to reconstruct the signal.”); and
determining a private dataset comprising a plurality of private data points from the plurality of private transform coefficients by applying, on the plurality of private transform coefficients, an inverse transform of the transform (¶ [0171], “At block 315, decoder 113 generates private data 140, as a differentially private output”; and Fig. 1, ¶ [0099], “FPA begins by computing a sequence Fk, comprising the first k Fourier coefficients in the DFT of Q(I). Then it perturbs Fk using the LPA algorithm with parameter to compute a noisy estimate {tilde over (F)}k. The perturbation may guarantee differential privacy. Finally, the algorithm computes the inverse DFT of PADn({tilde over (F)}k) to get {tilde over (Q)}k, an approximation to the original query answers Q(I).”);
3 ….
Amroabadi doesn’t disclose
1 …by applying an (ε, δ)-differential privacy mechanism…;
2 …, wherein the transform depends on the plurality of data points;
3 wherein the (ε, δ)-differential privacy mechanism is adapted such that the plurality of private data points is (ε, δ)-differential private.
Peterson, however, discloses
1 …by applying an (ε, δ)-differential privacy mechanism… (¶ [0095], “A further relaxation of the bounds is introduced by (ε, δ)-differential privacy that ensures that the ε-differential privacy bound holds with the probability δ.”);
3 wherein the (ε, δ)-differential privacy mechanism is adapted such that the plurality of private data points is (ε, δ)-differential private (¶¶ [0094]-[0095], “ This correction is commonly referred to as the noise introduced in the algorithm, its input, or output to ensure that the ε-differential privacy bound holds. A further relaxation of the bounds is introduced by (ε, δ)-differential privacy that ensures that the ε-differential privacy bound holds with the probability δ.”).
Regarding the combination of Amroabadi and Peterson, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the differential privacy system of Amroabadi to arrive at the claimed invention. KSR establishes that a rationale for obviousness is proven by showing a “use of [a] known technique to improve similar devices in the same way.” See MPEP § 2143(I)(C).
To substantiate the conclusion of obviousness under this KSR rationale, the Examiner finds pursuant to MPEP § 2143(I)(C):
1) the prior art contained a base system, namely the differential privacy system of Amroabadi, upon which the claimed invention can be seen as an “improvement” through the application of (ε, δ)-differential privacy;
2) the prior art contained a “comparable” system, namely the differential privacy system of Peterson, that has been improved in the same way as the claimed invention through the application of (ε, δ)-differential privacy; and
3) one of ordinary skill in the art could have applied the known improvement technique of applying the application of (ε, δ)-differential privacy to the base differential privacy system of Amroabadi, and the results would have been predictable to one of ordinary skill in the art.
Lopez, however, discloses
2 …, wherein the transform depends on the plurality of data points (¶ [0073], “The ideal transform is the multidimensional Karhunen-Loeve [data-dependent] transform, because it achieves complete de-correlation and maximum energy compaction.”);
Regarding the combination of Amroabadi-Peterson and Lopez, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the differential privacy system of Amroabadi-Peterson to arrive at the claimed invention. KSR establishes that a rationale for obviousness is proven by showing a “use of [a] known technique to improve similar devices in the same way.” See MPEP § 2143(I)(C).
To substantiate the conclusion of obviousness under this KSR rationale, the Examiner finds pursuant to MPEP § 2143(I)(C):
1) the prior art contained a base system, namely the differential privacy system of Amroabadi-Peterson, upon which the claimed invention can be seen as an “improvement” through the application of a data-dependent feature;
2) the prior art contained a “comparable” system, namely the transform system of Lopez, that has been improved in the same way as the claimed invention through the data-dependent feature; and
3) one of ordinary skill in the art could have applied the known improvement technique of applying the application of the data-dependent feature to the base differential privacy system of Amroabadi-Peterson, and the results would have been predictable to one of ordinary skill in the art.
Regarding Claim 6
Amroabadi in view of Peterson, and further in view of Lopez (“Amroabadi-Peterson-Lopez”) discloses the method according to claim 1, and Amroabadi further discloses
wherein the transform is an orthogonal transform (¶ [0007] of Applicant’s specification that is interpreted as admitted prior art: “It was shown that if an orthogonal transform is applied to each user's data ‘separately’”).
Regarding Claim 7
Amroabadi-Peterson-Lopez discloses the method according to claim 1, and Amroabadi further discloses
wherein for at least one of the transform coefficients, a variance value is determined from a probabilistic model assigned to the at least one of the transform coefficients (¶ [0078], “A Laplace Perturbation Algorithm (LPA) can add suitably-chosen noise to the true answers. The noise is generated according to the Laplace distribution. Lap(λ) can be denoted as a random variable drawn from the Laplace distribution. The distribution Lap(λ) has mean 0 and variance 2λ2 with the following density,…”; and ¶ [0099], “FPA begins by computing a sequence Fk, comprising the first k Fourier coefficients in the DFT of Q(I). Then it perturbs Fk using the LPA algorithm with parameter to compute a noisy estimate {tilde over (F)}k. The perturbation may guarantee differential privacy. ”).
Regarding Claim 8
Amroabadi-Peterson-Lopez discloses the method according to claim 1, and Amroabadi further discloses
further comprising:
setting at least one of the transform coefficients to zero such that a total sum of variance values of each of the plurality of transform coefficients is larger than a variance threshold (¶ [0199], “Only 25% of the [transform] coefficients are non-zero [the other being set to zero] (75% compression) which results in a significant privacy budget [as the total sum of variance values that are related to a variance threshold] saving.”).
Regarding Claim 9
Amroabadi-Peterson-Lopez discloses the method according to claim 1, and Amroabadi further discloses
wherein the following step is repeated as long as the total sum of variance values of the plurality of transform coefficients is larger than the variance threshold (¶ [0199]):
determining a low-variance transform coefficient of the plurality of transform coefficients which comprises a lowest nonzero variance value; and/or
setting the low-variance transform coefficient to zero (¶¶ [0139]-[0141], “Encoder 112 can be configured to present input data 110 in a sparse way, for example, represented with few large [low-variance] coefficients or elements, and lots of zeroes [being set], with a sparse domain.”).
Regarding Claim 10
Amroabadi-Peterson-Lopez discloses the method according to claim 1, and Amroabadi further discloses
wherein determining the plurality of private transform coefficients comprises adding independent noise to each non-zero transform coefficient of the plurality of transform coefficients (¶ [0154], ““In some embodiments, the added [independent] noise follows a Laplacian distribution, such as a Lap (k1/2Δ2(F)/ϵ) distribution, where k is the number of non-zero coefficients used to reconstruct the signal.””).
Regarding Claim 11
Amroabadi-Peterson-Lopez discloses the method according to claim 10, and Amroabadi further discloses
wherein the noise is Laplace distributed (¶ [0154], ““In some embodiments, the added [independent] noise follows a Laplacian distribution, such as a Lap (k1/2Δ2(F)/ϵ) distribution, where k is the number of non-zero coefficients used to reconstruct the signal.””).
Regarding Claim 12
Amroabadi-Peterson-Lopez discloses the method according to claim 1, and Peterson further discloses
wherein the (ε, δ)-differential privacy mechanism is adapted by adapting a noise distribution parameter associated with the (ε, δ)-differential privacy mechanism (¶ [0023], “In other words, c represents the upper bound for variance between the probabilities with which A generates an output from O. That effectively introduces a “correction” in algorithm A to ensure that the differential privacy bound holds for any two adjacent inputs. That correction [noise distribution parameter] is commonly referred to as noise introduced [added] in the algorithm, its input, or output to ensure that the ε-differential privacy bound holds. A further relaxation of the bounds is introduced by (ε, δ)-differential privacy [mechanism] that ensures that the ε-differential privacy bound holds with the probability δ.”).
Regarding the combination of Amroabadi and Peterson, the rationale to combine is the same as provided for claim 1 due to the overlapping subject matter of claims 1 and 12.
Regarding Claim 13
Amroabadi-Peterson-Lopez discloses the method according to claim 1, and Amroabadi further discloses
wherein the dataset comprises one data point per user of the plurality of users and correlations in the dataset consist of the inter-user correlations within the plurality of users (¶ [0022], “Time series stamping of individual client data [one data point per user] is a unique characteristic that makes existing (cross-sectional) privacy mechanisms inadequate, since successive timestamps [of data points involving different users, i.e., inter-users] from the same source can be highly correlated. This [inter-user] correlation can cause answers to different queries over time series data to also become correlated, e.g., a sequence of queries computing the average weight of a community at successive weeks.”).
Regarding Independent Claims 14 and 15
With respect to independent claims 14 and 15, a corresponding reasoning as given earlier for independent claim 1 applies, mutatis mutandis, to the subject matter of claims 14 and 15. Therefore, claims 14 and 15 are rejected, for similar reasons, under the grounds set forth for claim 1.
B. Claim 2 is rejected under 35 U.S.C. 103 as being unpatentable over Amroabadi in view of Peterson and Lopez, and further in view of Xu et al. (US 2009/0238484, “Xu”).
Regarding Claim 2
Amroabadi-Peterson-Lopez discloses the method according to claim 1, and Amroabadi further discloses
wherein the transform (¶ [0085]) is selected from…1
Amroabadi-Peterson-Lopez doesn’t disclose
1 …a set of transforms such that, when applying the transform on the plurality of data points, a decorrelation efficiency metric value of the plurality of {transform coefficients.
Xu, however, discloses
1 …a set of transforms such that, when applying the transform on the plurality of data points, a decorrelation efficiency metric value of the plurality of {transform coefficients (Amroabadi ¶ [0085)} is maximized (¶ [0014], “The resulting directional lapped transforms inherits the advantages of lapped transform, e.g., less blocking artifacts, better coding efficiency, lower computational complexity, etc. [] Unlike conventional lapped transforms [possibly selected], directional lapped transforms can efficiently de-correlate directional signals [data points]. Consequently, the DLT Coder described herein has advantages for uses such as image coding (2D signals), filtering of seismic data (3D signals), etc.”).
Regarding the combination of Amroabadi-Peterson-Lopez and Xu, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the differential privacy system of Amroabadi-Peterson-Lopez to arrive at the claimed invention. KSR establishes that a rationale for obviousness is proven by showing a “use of [a] known technique to improve similar devices in the same way.” See MPEP § 2143(I)(C).
To substantiate the conclusion of obviousness under this KSR rationale, the Examiner finds pursuant to MPEP § 2143(I)(C):
1) the prior art contained a base system, namely the differential privacy system of Amroabadi-Peterson-Lopez, upon which the claimed invention can be seen as an “improvement” through the application of a de-correlation feature;
2) the prior art contained a “comparable” system, namely the transform system of Xu, that has been improved in the same way as the claimed invention through the application of the de-correlation feature; and
3) one of ordinary skill in the art could have applied the known improvement technique of applying the application of the de-correlation feature to the base differential privacy system of Amroabadi-Peterson-Lopez, and the results would have been predictable to one of ordinary skill in the art.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to D'ARCY WINSTON STRAUB whose telephone number is (303)297-4405. The examiner can normally be reached Monday-Friday 9:00-5:00 Mountain Time.
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, WILLIAM KORZUCH can be reached at (571)272-7589. 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.
/D'Arcy Winston Straub/Primary Examiner, Art Unit 2491
Prosecution Insights