DETAILED ACTION
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 10/29/2025 has been entered.
Response to Arguments
The applicant’s Remarks pg. 10, filed 12/29/2025, regarding the status and amendments of the claims is hereby acknowledged.
The applicant’s Remarks pg. 10, filed 12/29/2025, regarding the summary of the telephone interview on 11/20/2025 is hereby acknowledged.
The applicant’s Remarks 10-11, filed 12/29/2025 regarding the rejection of claims 19-38 under 35 U.S.C. 112 have been fully considered. The examiner notes that the applicant’s arguments are directed to the newly amended limitations. Therefore, the examiner sets forth a new grounds of rejection in order to take into consideration the newly amended limitations.
With respect to the newly amended limitations, the applicant argues the following:
The Office Action states that paragraphs [0109] and [0138]-[0141] of Hoctor disclose that, in Hoctor, "binge-watching takes into consideration times when the viewer uses another service to view content." However, considering whether a user uses another service to view content differs from the features of Applicant's amended claims, which determine that content being binge-consumed via a first content source was interrupted by different content from the same first content source.
Pages 5-6 of the Office Action cite paragraphs [0124] and [0121]-[0122] of Hoctor and alleges that based on the teachings of those paragraphs, "a person of ordinary skill in the art would have reasonably inferred that the recommendation described in Hoctor paragraph 115 is based on collaborative information of other viewers that have watched the same content and different content."
The examiner respectfully disagrees. In response to applicant’s, the test for obviousness is not whether the features of a secondary reference may be bodily incorporated into the structure of the primary reference; nor is it that the claimed invention must be expressly suggested in any one or all of the references. Rather, the test is what the combined teachings of the references would have suggested to those of ordinary skill in the art. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981). Additionally, on the issue of obviousness, the Supreme Court stated the analysis of a rejection on obviousness grounds need not seek out precise teachings directed to the specific subject matter of the challenged claim, for a court can take account of the inferences and creative steps that a person of ordinary skill in the art would employ. See KSR International Co. v. Teleflex Inc., 550 U.S. 398, 418, 82 USPQ2d 1385 (2007). The obvious analysis cannot be confined by a formalistic conception of the words teaching, suggestion, and motivation. Id. at 419. Further, the Court stated that common sense teaches, however, that familiar items may have obvious uses beyond their primary purposes, and in many cases a person of ordinary skill will be able to fit the teachings of multiple patents together like pieces of a puzzle. Id. at 420.
With respect to the teachings of Hoctor, the prior art teaches tracking when the viewer is binge watching a series of episodes (para 142) in order to recommend additional content available to the user (para 154). More importantly, in additional to tracking the binge watching activities of a particular user, Hoctor also compares a first user’s watching history to other users (para 138-142). The teachings of Hoctor dovetail with the additional prior art of record as will be discussed below.
Furthermore, whereas the applicant has amended the independent claims to recite a first, second, and third binge-consumption of a first and second user, it is important to note that the combination of prior art of record tracks every users binge-consumption. Additionally, the applicant has amended the claims to recite interruptions to each of the binge-consumption, it is important to note that the teachings of the prior art of record also teaches tracking every user’s viewing history and also discloses what elements are taking into consideration in a viewing history (e.g., watching different content than the media series being consumed. Equally important is that Vasileva also recognizes that binge-watching includes situations wherein a viewer watches several episodes without any intermission which suggests that the viewer is not switching to other content other than the binging content (Vasileva pg. 2 “binge-watching and the need to see several episodes of a showing without intermission”). A person of ordinary skill in the art would reasonably infer predictions, the other users that are similarly situated. All things considered, the examiner will rely on the prior art of record in order to address the newly amended limitations.
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 § 103
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 19-38 are rejected under 35 U.S.C. 103 as being unpatentable over Hoctor et al., US 2015/0312609 (hereafter Hoctor) and in further view of Vasileva, Konstantina, How Netflix uses data to keep you binge-watching & personalize your viewing experience, December 9, 2017, The Data Nudge, pg. 1-9 (hereafter Vasileva) and in further view of Hari Haran et al., US20200226493A1 (hereafter Hari Haran) and in further view of Dykstra et al., US 8554640B1 (hereafter Dykstra) and in further view of Errico; James H. et al. US 20060282856 A1 (hereafter Errico)and in further view of Shen; Qifeng US 10341703 B1 (hereafter Shen).
Regarding claim 19, “a method comprising: identifying a first content consumption history of a first user; determining that the first content consumption history indicates that, during a first time period that is less than or equal to a threshold amount of time: first binge-consumption of at least one season of a first content series by way of a first content source began; a first interruption occurred in which consumption of different content than the first content series, by way of the first content source, interrupted the first binge-consumption; and the first binge-consumption was completed after the first interruption; based at least in part on the determining, identifying a second user associated with a second content consumption history indicating that, during a second time period that is less than or equal to the threshold amount of time: second binge-consumption of at least one season of the first content series by way of the first content source began; a second interruption occurred in which consumption of different content than the first content series, by way of the first content source, interrupted the second binge-consumption; and the second binge-consumption was completed after the second interruption; based at least in part on determining a correspondence between (i) the first interruption and subsequent completion of the first binge-consumption and (ii) the second interruption and subsequent completion of the second binge-consumption: determining that the second consumption history indicates that, during a third time period that is less than or equal to the threshold amount of time: third binge-consumption of at least one season of a second content series, different from the first content series, began, by way of the first content source; a third interruption in which consumption of different content than the second content series, by way of the first content source, interrupted the third binge-consumption; and the third binge-consumption was completed after the third interruption; determining that the first user is likely to be interested in the second content series based on a similarity of (iii) the third interruption and subsequent completion of the third binge consumption to (i) the first interruption and subsequent completion of the first binge consumption; and generating for display, on a device of the first user, a recommendation to begin consumption of the second content series” Hoctor para 115 teaches recommendation of content associated with preventing a user from termination a service and in response to determining that a user has consumed all the media assets in a particular series, the media guidance application may recommend media assets in a different series; Hoctor teaches content consumption history indicates that consumption of a first content series has been completed during a first time period that is less than or equal to a threshold amount of time wherein Hoctor para 138-142 determining a viewing history of a rate at which the user has consumed the series of media assets over a time period comprising when a viewer is watching episodes of a series back-to-back without a break in between episodes and also compares the rate of the user to threshold rates; More importantly, in additional to tracking the binge watching activities of a particular user, Hoctor also compares a first user’s watching history to other users (para 138-142) wherein binge-watching takes into consideration times when the viewer uses another service to view content – See 109 and 138-141. Whereas claim 1 recites a first, second, and third binge-consumption of a first and second user, it is important to note that the combination of prior art of record tracks every users binge-consumption. Additionally, claim 1 recites interruptions to each of the binge-consumption, it is important to note that the teachings of the prior art of record also teaches tracking every user’s viewing history and also discloses what elements are taking into consideration in a viewing history (e.g., watching different content than the media series being consumed. See also Hoctor para 48 teaches the rate of consumption of the user may correspond to an amount of content consumed (e.g., a number of media assets, a number of hours during which the user viewed programming, a number of episodes, season, series, etc., and/or any other measure of content consumed) over a period of time (e.g., hours, days, weeks, etc.) to a threshold rate (e.g., in the same units of measure); Hoctor para 50-52 teaches behavioral characteristics comprising “a viewing history” refers to a collection of information related to media content consumed by a user. For example, the viewing history may indicate an amount of media content consumed by the user, the rate of media content consumed by the user, particular times when a user consumed media content, user preferences for media content consumed by the user, and/or any other information related to the consumption of media content by the user. In some embodiments, the viewing history may indicate user preference or characteristics of media content preferred by the user. As used herein, a “characteristic” refers to any attribute of a media asset, series of media asset, and/or type of media asset that distinguishes the media asset, series of media asset, and/or type of media asset from other media assets, series, and/or types.
Regarding “based at least in part on the determining, identifying a second user associated with a second content consumption history indicating that, during a second time period that is less than or equal to the threshold amount of time: second binge-consumption of at least one season of the first content series by way of the first content source began; a second interruption occurred in which consumption of different content than the first content series, by way of the first content source, interrupted the second binge-consumption; and the second binge-consumption was completed after the second interruption; based at least in part on determining a correspondence between (i) the first interruption and subsequent completion of the first binge-consumption and (ii) the second interruption and subsequent completion of the second binge-consumption: determining that the second consumption history indicates that, during a third time period that is less than or equal to the threshold amount of time: third binge-consumption of at least one season of a second content series, different from the first content series, began, by way of the first content source; a third interruption in which consumption of different content than the second content series, by way of the first content source, interrupted the third binge-consumption; and the third binge-consumption was completed after the third interruption; determining that the first user is likely to be interested in the second content series based on a similarity of (iii) the third interruption and subsequent completion of the third binge consumption to (i) the first interruption and subsequent completion of the first binge consumption; and generating for display, on a device of the first user, a recommendation to begin consumption of the second content series” as discussed above, in additional to tracking the binge watching activities of a particular user, Hoctor also compares a first user’s watching history to other users and related thresholds (para 138-142); see also Hoctor para 115 teaches in response to determining that a user has consumed all the media assets in a particular series, the media guidance application may recommend media assets in a different series as disclosed in Fig. 5; Hoctor teaches that in order to display recommendations as discussed in para 115 and Fig. 5, Hoctor para 121-122 predictive attributes engine 620 related to recommended content comprises one or more of the data it receives to generate attributes that represent a population of users with similar activity (corresponds to based at least in part on the determining, identifying a second user associated with a second content consumption history); Hoctor para 124 Learning model engine 640 may perform the same analysis for each other user for which data is available (e.g., in a database for historical outcomes 624) and who terminated access to the same source or service. Hoctor para 124 teaches a learning model performs an analysis for each other user wherein para 121-122 disclosing the analysis is performed on viewing history, therefore, based on the teachings of Hoctor, a person of ordinary skill in the art would have reasonably inferred that the recommendation described in Hoctor paragraph 115 is based on the collaborative information of other viewers that have watched the same content and different content. However, Hoctor does not explicitly reference (i.e., in haec verba) generating for display, on a device of the first user, a recommendation to begin consumption of the second content series, however, a person of ordinary skill in the art would reasonably infer, based on Hoctor’s teachings discussed supra, that the viewing habits of all viewers, including binge watched shows, are included as part of the predictive attributes engine as disclosed in para 121-122, 124 because Hoctor analyzes predictive attributes engine 620 related to recommended content comprises one or more of the data it receives to generate attributes that represent a population of users with similar activity. Hoctor does not use the same terms for disclosing wherein the first content consumption history further indicates first behavioral information in relation to consumption of different content than the first content series, during the first time period.
The inferences drawn in Hoctor are further evidenced in an analogous art to Vasileva disclosing how one particular content provider utilizes metrics for all users in order to predict additional viewing content based on metrics comprising binge-watching data (pg. 7-9 e.g., “…[m]etrics include time gaps between viewing show episodes, preferred times for watching content, preferred devices, rates of pausing or leaving certain types of content, ratings, search history, even in movie characteristics like colors, volume, etc. The point of understanding all this data? To learn what factors keep viewers interested and less likely to cancel viewing…” in addition to binge watching data.) Vasileva teaches analyzing a first viewer data comprising binge-watching data and collective data for other users in order to make recommendations to the first viewer because the prior art recognizes utilizing viewer analytics for millions of users in order to build predictions and relates to “determining that the first user is likely to be interested in the second content series based on a similarity of (iii) the third interruption and subsequent completion of the third binge consumption to (i) the first interruption and subsequent completion of the first binge consumption; and generating for display, on a device of the first user, a recommendation to begin consumption of the second content series.” Vasileva also recognizes that binge-watching includes situations wherein a viewer watches several episodes without any intermission which suggests that the viewer is not switching to other content other than the binging content (Vasileva pg. 2 “binge-watching and the need to see several episodes of a showing without intermission”). A person of ordinary skill in the art would reasonably infer predictions, the other users that are similarly situated (i.e., determining that the first user is likely to be interested in the second content series based on a similarity of (iii) the third interruption and subsequent completion of the third binge consumption to (i) the first interruption and subsequent completion of the first binge consumption; and generating for display, on a device of the first user, a recommendation to begin consumption of the second content series.) If the purpose is to utilize and analyze binge data, a person of ordinary skill would appreciate the benefit of metrics as taught by the prior art (e.g., Hoctor). Therefore, merely combining viewing metrics for one user or a plurality of users in a database similar to Netflix would not only allow a content provider to analyze and utilizing a single user’s viewing preferences for making recommendation predications, but also make a recommendation, on what to watch next, to the single user after said user binge watches a complete season (or available episodes of a season) based on metrics for all users and what content other users watched, and enjoyed to completion, after watching the same season as the first binge watched show.
With respect to determining that the first user is likely to be interested in the second content series based on a similarity of (iii) the third interruption and subsequent completion of the third binge consumption to (i) the first interruption and subsequent completion of the first binge consumption; and generating for display, on a device of the first user, a recommendation to begin consumption of the second content series, Hari Haran teaches a method for training a similarity mode used to predict similarity between media content in order to provide a better viewing experience by enabling the recommendation of more relevant items (para 29); Hari Haran para 74 – if a person binge watches a TV series, it is fair to say they liked the TV series; para 14 From a computational perspective, items are often considered similar if they are referenced by the same items. Both CB and CF used for predicting recommendations can also be used to predict similarity by analyzing these references. CB would consider two items similar if it references the same metadata, and CF would consider two items similar if it references the same users. For example, in a VoD system two movies could be considered similar by CB if they share the same genre, director, and cast. CF would consider two items similar if they are both consumed, liked, and/or disliked by same users, for example, if every user who enjoyed movie A also enjoyed movie B, then the system can infer that movie A and movie B somehow must be similar to one another. The CF approach to similarity as used by Amazon has been demonstrated to be useful in academia and industry, see G. Linden et. al. “Amazon.com Recommendations: Item-to-Item Collaborative Filtering” IEEE Internet Computing Industry Report, January-February 2003 (the contents of which are incorporated herein by reference). A person of ordinary skill in the art would reasonably infer that collective filtering enables a recommendation system to recommend content to a first user based on the viewing history of other users who have viewed the same media content as the first user in further view of Dykstra.
Hari Haran, similar to Vasileva and Hoctor, renders obvious the limitations when viewed in light of Dykstra’s invention for content completion recommendation wherein content comprises media content comprising episodes (Abstract, col. 1:61-67 to col. 2:1-27) and further teaches that a user will be provided a recommendation of a second content after the content completion of a first content based on whether other viewers which liked a particular first content also enjoyed the second content (col. 12:61-67 to col. 13:1-42). Therefore, whereas Dykstra teaches that a user will be provided a recommendation of a second content after the content completion of a first content based on whether other viewers which liked a particular first content also enjoyed the second content. Dykstra does not explicitly disclose a user switching or changing between different content items.
The motivation to modify Hoctor, Vasileva, and Dykstra is further evidenced by Errico disclosing a collaborative recommendation system for recommending programming content to a viewer comprising recommending a program series which the viewer has not previously viewed based on the recommended program series having been viewed by other viewers with similar interests (para 142-143, 162) and further teaches “Special icons may be used to designate negative preference wherever recommendations are displayed (e.g. the "Friends Recommend" display). [0170] Endorser status--description of how the item is currently being consumed by the recommending entities. This may include indication that viewers are currently watching or have stopped watching. [0171] Endorser supplementary data available--description of data and services that may be available from the recommending entities.” (Errico para 163-171; see also para 177-180 disclosing the display of recommended content along with indicators identifying the behavioral attributes of the at least one other user relating to whether the at least one other user viewed the program content and whether the at least one other user would recommend the content comprising negative degrees). Errico is silent with respect to the deficiency of Hoctor, Vasileva, Hari Haran, and Dykstra wherein the first content consumption history further indicates binge-consumption interruptions in relation to consumption of different content than the first content series, during the first time period.
Examiner’s Note: See prior art of record cited but not relied upon Carlberg; Marvin Charles et al. US 20090060468 A1 (hereafter Carlberg) to avoid duplicative references wherein Fig. 15 and para [0030-0032] of Carlberg teaches that when providing a recommendation of content viewed by different viewers; See also RUIZ-VELASCO; ENRIQUE et al. US 20090019488 A1 (hereafter Ruiz-Velasco) Fig. 2C and 3D and [0037] providing embodiments comprising content rating and comments made by different viewers while watching the recommended content.
In an analogous art, Shen, similar to Hoctor, Vasileva, Hari Haran, Dykstra, and Errico, teaches analyzing viewing history but Shen further teaches tracking a user’s viewing history utilizing a user profile of a viewer comprising tracking attributes of viewers including viewing habits such as tendency to binge-watch, a channel changing tendency, and a propensity to watch a particular show (col. 4:13-67).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Hoctor’s analyze a first user’s viewing history comprising when a viewer has binge-watched every available episode of a particular show and using collective viewing data of different viewers to make recommendations to the first viewer of what episode/content to watch next by further incorporating known elements of Vasileva’s invention for analyzing binge watching data of each show for every viewer and further generates a new show series recommendation based on a collective analysis of all viewers determined interests in order to provide recommendations, to viewers that have finished watching a particular series of content, of additional binge content that is likely to be enjoyed by the viewer, based on metrics from other viewers that have already completed the similar episode/series/season because the selections are based on content that is popular amongst a plurality of viewers as taught by Vasileva. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Hoctor and Vasileva for analyzing viewer binge watching history to make recommendations to other viewers of what programs to watch next by further incorporating known elements of Hari Haran’s invention for utilizing collaborative consumption/watching data in order to determine a new recommendation after the completion of a first media content based on whether a collective group of viewers have also watched the same first media content and also enjoyed a second different media content because the prior art recognizes the use of collaborative consumption in order to make content completion recommendations as taught by Dykstra. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Hoctor, Vasileva, Hari Haran, and Dykstra for analyzing viewer viewing history comprising binge watching history to make recommendations to other viewers of what programs to watch next by further incorporating known elements of Errico’s invention for collaborative recommendation system utilized for recommending programming content to a viewer comprising a program series which the viewer has not previously viewed based on the recommended program series having been viewed by other viewers with similar interests and further display the recommended content along with a plurality of indicators identifying the behavioral attributes of the at least one other user relating to whether the at least one other identified user viewed the program content and whether the at least one other identified user would recommend the content comprising negative degrees in order to provide the viewer with data which identifies additional factors to consider viewing the recommended content. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Hoctor, Vasileva, Hari Haran, Dykstra, and Errico for analyzing viewer viewing history comprising binge watching history and collaborative data of other viewers with similar attributes to make additional recommendations to other viewers of what programs to watch next by further incorporating known elements of Shen’s invention for tracking each users’ viewing history utilizing a user profile of a viewer comprising tracking attributes of viewers including viewing habits such as tendency to binge-watch, a channel changing tendency, and a propensity to watch a particular show in order to improve the accuracy of the prior art that provides targeted content recommendation to viewers based on similar viewing history comprising attributes related to binge-watching in combination with changing channels data in order to identify the level of interest in a program episode as the mere combination of known attributes is likely to lead to predictable results when creating user profiles relating to viewing history.
Regarding claim 20, “wherein the generating for display the recommendation to begin the consumption of the second content series comprises determining that, for each respective user of a threshold number of users, the respective user is associated with a content consumption history indicating: during a time period that is less than or equal to the threshold amount of time, the first content series was completed” is further rejected on obviousness grounds as discussed in the rejection of claim 1 wherein Hoctor para 122 teaches Predictive attributes engine 620 processes one or more of the data it receives to generate attributes that represent a population of users with similar activity.
Regarding claim 21, “further comprising: determining a number of users that have binge-watched the second content series, wherein recommending the second content series for the first user to begin consumption is based at least in part on the determined number of users that have binge-watched the second content series.” is further rejected on obviousness grounds as discussed in the rejection of claim 1 wherein the claim limitation is a combination of the teachings of Hoctor para 115 in response to determining that a user has consumed all the media assets in a particular series, the media guidance application may recommend media assets in a different series; para 138-139 media guidance application may also determine if episodes of a series are watched continuously (e.g., back-to-back without a break in between episodes) and the rate at which content is consumed.
Regarding claim 22, “further comprising: analyzing a number of users that have binge-watched one or more content series; identifying a most commonly consumed content series of the one or more content series based at least in part on a number of users that have binge-watched the one or more content series; sorting the one or more second content series in an order of popularity based at least in part on the number of users that have binge-watched the one or more content series, and wherein the recommendation to begin consumption of the content series is further based at least in part on the most commonly consumed content series.” is further rejected on obviousness grounds as discussed in the rejection of claim 19-21 wherein Hoctor para 138-139 media guidance application may also determine if episodes of a series are watched continuously (e.g., back-to-back without a break in between episodes) and the rate at which content is consumed and see also Errico disclosing a collaborative recommendation system for recommending programming content to a viewer comprising recommending a program series which the viewer has not previously viewed based on the recommended program series having been viewed by other viewers with similar interests (para 142-143, 162) and further teaches “Special icons may be used to designate negative preference wherever recommendations are displayed (e.g. the "Friends Recommend" display). [0170] Endorser status--description of how the item is currently being consumed by the recommending entities. This may include indication that viewers are currently watching or have stopped watching. [0171] Endorser supplementary data available--description of data and services that may be available from the recommending entities.” (Errico para 163-171; see also para 177-180 disclosing the display of recommended content along with indicators identifying the behavioral attributes of the at least one other user relating to whether the at least one other user viewed the program content and whether the at least one other user would recommend the content comprising negative degrees).
Examiner’s Note: See prior art of record cited but not relied upon Carlberg; Marvin Charles et al. US 20090060468 A1 (hereafter Carlberg) to avoid duplicative references wherein Fig. 15 and para [0030-0032] of Carlberg teaches that when providing a recommendation of content viewed by different viewers; See also RUIZ-VELASCO; ENRIQUE et al. US 20090019488 A1 (hereafter Ruiz-Velasco) Fig. 2C and 3D and [0037] providing embodiments comprising content rating and comments made by different viewers while watching the recommended content.
Regarding claim 23, “wherein the sorting the one or more content series comprises: identifying one or more service providers respectively associated with the one or more content series; sorting the one or more content series based at least in part on service providers of the one or more content series and wherein the recommendation to begin consumption of the content series is further based at least in part on determining a most common service provider of the one or more content series” is further rejected on obviousness grounds as discussed in the rejection of claim 19-22 wherein Errico further teaches para 23 the recommendations may be provided by the personal video recorder or otherwise from a service provider.
Regarding claim 24, “wherein the determining that the first content binge-consumption history indicates that the consumption of the first content series by way of a first content source has been completed during the first time period comprises determining that the first user has consumed at least a threshold number of episodes of the first content series within the threshold amount of time” is further rejected on obviousness grounds as discussed in the rejection of claim 19-23 wherein Hoctor para 115 teaches recommendation of content associated with preventing a user from termination a service and in response to determining that a user has consumed all the media assets in a particular series, the media guidance application may recommend media assets in a different series; Hoctor teaches content consumption history indicates that consumption of a first content series has been completed during a first time period that is less than or equal to a threshold amount of time wherein Hoctor para 138-142 determining a viewing history of a rate at which the user has consumed the series of media assets over a time period comprising when a viewer is watching episodes of a series back-to-back without a break in between episodes and also compares the rate of the user to threshold rates. Hoctor para 48 teaches the rate of consumption of the user may correspond to an amount of content consumed (e.g., a number of media assets, a number of hours during which the user viewed programming, a number of episodes, season, series, etc., and/or any other measure of content consumed) over a period of time (e.g., hours, days, weeks, etc.) to a threshold rate (e.g., in the same units of measure); Hoctor para 50-52 teaches behavioral characteristics comprising “a viewing history” refers to a collection of information related to media content consumed by a user. For example, the viewing history may indicate an amount of media content consumed by the user, the rate of media content consumed by the user, particular times when a user consumed media content, user preferences for media content consumed by the user, and/or any other information related to the consumption of media content by the user. Hoctor teaches that in order to display recommendations as discussed in para 115 and Fig. 5, Hoctor para 121-122 predictive attributes engine 620 related to recommended content comprises one or more of the data it receives to generate attributes that represent a population of users with similar activity. See also Vasileva teaches analyzing a first viewer data comprising binge-watching data and collective data for other users in order to make recommendations to the first viewer because the prior art recognizes utilizing viewer analytics for millions of users in order to build predictions. See also Hari Haran teaches a method for training a similarity mode used to predict similarity between media content in order to provide a better viewing experience by enabling the recommendation of more relevant items (para 29); Hari Haran para 74 – if a person binge watches a TV series, it is fair to say they liked the TV series; para 14 From a computational perspective, items are often considered similar if they are referenced by the same items. Both CB and CF used for predicting recommendations can also be used to predict similarity by analyzing these references. CB would consider two items similar if it references the same metadata, and CF would consider two items similar if it references the same users. For example, in a VoD system two movies could be considered similar by CB if they share the same genre, director, and cast. CF would consider two items similar if they are both consumed, liked, and/or disliked by same users, for example, if every user who enjoyed movie A also enjoyed movie B, then the system can infer that movie A and movie B somehow must be similar to one another. The CF approach to similarity as used by Amazon has been demonstrated to be useful in academia and industry, see G. Linden et. al. “Amazon.com Recommendations: Item-to-Item Collaborative Filtering” IEEE Internet Computing Industry Report, January-February 2003 (the contents of which are incorporated herein by reference). A person of ordinary skill in the art would reasonably infer that collective filtering enables a recommendation system to recommend content to a first user based on the viewing history of other users who have viewed the same media content as the first user.
Regarding claim 25, “wherein the identifying the second user comprises determining that the second user has binge-watched at least a threshold number of episodes of the first content series within the threshold amount of time” is further rejected on obviousness grounds as discussed in the rejection of claim 19-24 wherein Hoctor para 115 teaches recommendation of content associated with preventing a user from termination a service and in response to determining that a user has consumed all the media assets in a particular series, the media guidance application may recommend media assets in a different series; Hoctor teaches content consumption history indicates that consumption of a first content series has been completed during a first time period that is less than or equal to a threshold amount of time wherein Hoctor para 138-142 determining a viewing history of a rate at which the user has consumed the series of media assets over a time period comprising when a viewer is watching episodes of a series back-to-back without a break in between episodes and also compares the rate of the user to threshold rates. Hoctor para 48 teaches the rate of consumption of the user may correspond to an amount of content consumed (e.g., a number of media assets, a number of hours during which the user viewed programming, a number of episodes, season, series, etc., and/or any other measure of content consumed) over a period of time (e.g., hours, days, weeks, etc.) to a threshold rate (e.g., in the same units of measure); Hoctor para 50-52 teaches behavioral characteristics comprising “a viewing history” refers to a collection of information related to media content consumed by a user. For example, the viewing history may indicate an amount of media content consumed by the user, the rate of media content consumed by the user, particular times when a user consumed media content, user preferences for media content consumed by the user, and/or any other information related to the consumption of media content by the user. Hoctor teaches that in order to display recommendations as discussed in para 115 and Fig. 5, Hoctor para 121-122 predictive attributes engine 620 related to recommended content comprises one or more of the data it receives to generate attributes that represent a population of users with similar activity. See also Vasileva teaches analyzing a first viewer data comprising binge-watching data and collective data for other users in order to make recommendations to the first viewer because the prior art recognizes utilizing viewer analytics for millions of users in order to build predictions. See also Hari Haran teaches a method for training a similarity mode used to predict similarity between media content in order to provide a better viewing experience by enabling the recommendation of more relevant items (para 29); Hari Haran para 74 – if a person binge watches a TV series, it is fair to say they liked the TV series; para 14 From a computational perspective, items are often considered similar if they are referenced by the same items. Both CB and CF used for predicting recommendations can also be used to predict similarity by analyzing these references. CB would consider two items similar if it references the same metadata, and CF would consider two items similar if it references the same users. For example, in a VoD system two movies could be considered similar by CB if they share the same genre, director, and cast. CF would consider two items similar if they are both consumed, liked, and/or disliked by same users, for example, if every user who enjoyed movie A also enjoyed movie B, then the system can infer that movie A and movie B somehow must be similar to one another. The CF approach to similarity as used by Amazon has been demonstrated to be useful in academia and industry, see G. Linden et. al. “Amazon.com Recommendations: Item-to-Item Collaborative Filtering” IEEE Internet Computing Industry Report, January-February 2003 (the contents of which are incorporated herein by reference). A person of ordinary skill in the art would reasonably infer that collective filtering enables a recommendation system to recommend content to a first user based on the viewing history of other users who have viewed the same media content as the first user.
Regarding claim 26, “wherein the identifying the second user comprises analyzing viewing data derived from behavioral attributes of the second user and determining that the viewing data indicates that preferences of the second user are related to preferences of the first user” is further rejected on obviousness grounds as discussed in the rejection of claim 19-25 wherein Hoctor para 115 teaches recommendation of content associated with preventing a user from termination a service and in response to determining that a user has consumed all the media assets in a particular series, the media guidance application may recommend media assets in a different series; Hoctor teaches content consumption history indicates that consumption of a first content series has been completed during a first time period that is less than or equal to a threshold amount of time wherein Hoctor para 138-142 determining a viewing history of a rate at which the user has consumed the series of media assets over a time period comprising when a viewer is watching episodes of a series back-to-back without a break in between episodes and also compares the rate of the user to threshold rates. Hoctor para 48 teaches the rate of consumption of the user may correspond to an amount of content consumed (e.g., a number of media assets, a number of hours during which the user viewed programming, a number of episodes, season, series, etc., and/or any other measure of content consumed) over a period of time (e.g., hours, days, weeks, etc.) to a threshold rate (e.g., in the same units of measure); Hoctor para 50-52 teaches behavioral characteristics comprising “a viewing history” refers to a collection of information related to media content consumed by a user. For example, the viewing history may indicate an amount of media content consumed by the user, the rate of media content consumed by the user, particular times when a user consumed media content, user preferences for media content consumed by the user, and/or any other information related to the consumption of media content by the user. Hoctor teaches that in order to display recommendations as discussed in para 115 and Fig. 5, Hoctor para 121-122 predictive attributes engine 620 related to recommended content comprises one or more of the data it receives to generate attributes that represent a population of users with similar activity. See also Vasileva teaches analyzing a first viewer data comprising binge-watching data and collective data for other users in order to make recommendations to the first viewer because the prior art recognizes utilizing viewer analytics for millions of users in order to build predictions. See also Hari Haran teaches a method for training a similarity mode used to predict similarity between media content in order to provide a better viewing experience by enabling the recommendation of more relevant items (para 29); Hari Haran para 74 – if a person binge watches a TV series, it is fair to say they liked the TV series; para 14 From a computational perspective, items are often considered similar if they are referenced by the same items. Both CB and CF used for predicting recommendations can also be used to predict similarity by analyzing these references. CB would consider two items similar if it references the same metadata, and CF would consider two items similar if it references the same users. For example, in a VoD system two movies could be considered similar by CB if they share the same genre, director, and cast. CF would consider two items similar if they are both consumed, liked, and/or disliked by same users, for example, if every user who enjoyed movie A also enjoyed movie B, then the system can infer that movie A and movie B somehow must be similar to one another. The CF approach to similarity as used by Amazon has been demonstrated to be useful in academia and industry, see G. Linden et. al. “Amazon.com Recommendations: Item-to-Item Collaborative Filtering” IEEE Internet Computing Industry Report, January-February 2003 (the contents of which are incorporated herein by reference). A person of ordinary skill in the art would reasonably infer that collective filtering enables a recommendation system to recommend content to a first user based on the viewing history of other users who have viewed the same media content as the first user.
Regarding claim 27, “wherein the behavioral attributes of the second user comprise usage data indicating a number of episodes the second user continuously consumed in at least one consumption session and an amount of time for the second user to complete the consumption of the number of episodes” is further rejected on obviousness grounds as discussed in the rejection of claim 19-25 wherein Hoctor para 115 teaches recommendation of content associated with preventing a user from termination a service and in response to determining that a user has consumed all the media assets in a particular series, the media guidance application may recommend media assets in a different series; Hoctor teaches content consumption history indicates that consumption of a first content series has been completed during a first time period that is less than or equal to a threshold amount of time wherein Hoctor para 138-142 determining a viewing history of a rate at which the user has consumed the series of media assets over a time period comprising when a viewer is watching episodes of a series back-to-back without a break in between episodes and also compares the rate of the user to threshold rates. Hoctor para 48 teaches the rate of consumption of the user may correspond to an amount of content consumed (e.g., a number of media assets, a number of hours during which the user viewed programming, a number of episodes, season, series, etc., and/or any other measure of content consumed) over a period of time (e.g., hours, days, weeks, etc.) to a threshold rate (e.g., in the same units of measure); Hoctor para 50-52 teaches behavioral characteristics comprising “a viewing history” refers to a collection of information related to media content consumed by a user. For example, the viewing history may indicate an amount of media content consumed by the user, the rate of media content consumed by the user, particular times when a user consumed media content, user preferences for media content consumed by the user, and/or any other information related to the consumption of media content by the user. Hoctor teaches that in order to display recommendations as discussed in para 115 and Fig. 5, Hoctor para 121-122 predictive attributes engine 620 related to recommended content comprises one or more of the data it receives to generate attributes that represent a population of users with similar activity. See also Vasileva teaches analyzing a first viewer data comprising binge-watching data and collective data for other users in order to make recommendations to the first viewer because the prior art recognizes utilizing viewer analytics for millions of users in order to build predictions. See also Hari Haran teaches a method for training a similarity mode used to predict similarity between media content in order to provide a better viewing experience by enabling the recommendation of more relevant items (para 29); Hari Haran para 74 – if a person binge watches a TV series, it is fair to say they liked the TV series; para 14 From a computational perspective, items are often considered similar if they are referenced by the same items. Both CB and CF used for predicting recommendations can also be used to predict similarity by analyzing these references. CB would consider two items similar if it references the same metadata, and CF would consider two items similar if it references the same users. For example, in a VoD system two movies could be considered similar by CB if they share the same genre, director, and cast. CF would consider two items similar if they are both consumed, liked, and/or disliked by same users, for example, if every user who enjoyed movie A also enjoyed movie B, then the system can infer that movie A and movie B somehow must be similar to one another. The CF approach to similarity as used by Amazon has been demonstrated to be useful in academia and industry, see G. Linden et. al. “Amazon.com Recommendations: Item-to-Item Collaborative Filtering” IEEE Internet Computing Industry Report, January-February 2003 (the contents of which are incorporated herein by reference). A person of ordinary skill in the art would reasonably infer that collective filtering enables a recommendation system to recommend content to a first user based on the viewing history of other users who have viewed the same media content as the first user.
Regarding claim 28, “wherein the viewing data derived from the behavioral attributes of the second user indicates at least one of: types of content the second user typically consumes, times of day the second user typically consumes content, whether the second user interacts with a social network, typical times the second user interacts with the social network to post information, whether the second user typically consumes free or paid content, subscription information associated with the second user, or brain activity information associated with the second user” is further rejected on obviousness grounds as discussed in the rejection of claim 19-27 wherein Hoctor para 115 teaches recommendation of content associated with preventing a user from termination a service and in response to determining that a user has consumed all the media assets in a particular series, the media guidance application may recommend media assets in a different series; Hoctor teaches content consumption history indicates that consumption of a first content series has been completed during a first time period that is less than or equal to a threshold amount of time wherein Hoctor para 138-142 determining a viewing history of a rate at which the user has consumed the series of media assets over a time period comprising when a viewer is watching episodes of a series back-to-back without a break in between episodes and also compares the rate of the user to threshold rates. Hoctor para 48 teaches the rate of consumption of the user may correspond to an amount of content consumed (e.g., a number of media assets, a number of hours during which the user viewed programming, a number of episodes, season, series, etc., and/or any other measure of content consumed) over a period of time (e.g., hours, days, weeks, etc.) to a threshold rate (e.g., in the same units of measure); Hoctor para 50-52 teaches behavioral characteristics comprising “a viewing history” refers to a collection of information related to media content consumed by a user. For example, the viewing history may indicate an amount of media content consumed by the user, the rate of media content consumed by the user, particular times when a user consumed media content, user preferences for media content consumed by the user, and/or any other information related to the consumption of media content by the user. Hoctor teaches that in order to display recommendations as discussed in para 115 and Fig. 5, Hoctor para 121-122 predictive attributes engine 620 related to recommended content comprises one or more of the data it receives to generate attributes that represent a population of users with similar activity. See also Vasileva teaches analyzing a first viewer data comprising binge-watching data and collective data for other users in order to make recommendations to the first viewer because the prior art recognizes utilizing viewer analytics for millions of users in order to build predictions. See also Hari Haran teaches a method for training a similarity mode used to predict similarity between media content in order to provide a better viewing experience by enabling the recommendation of more relevant items (para 29); Hari Haran para 74 – if a person binge watches a TV series, it is fair to say they liked the TV series; para 14 From a computational perspective, items are often considered similar if they are referenced by the same items. Both CB and CF used for predicting recommendations can also be used to predict similarity by analyzing these references. CB would consider two items similar if it references the same metadata, and CF would consider two items similar if it references the same users. For example, in a VoD system two movies could be considered similar by CB if they share the same genre, director, and cast. CF would consider two items similar if they are both consumed, liked, and/or disliked by same users, for example, if every user who enjoyed movie A also enjoyed movie B, then the system can infer that movie A and movie B somehow must be similar to one another. The CF approach to similarity as used by Amazon has been demonstrated to be useful in academia and industry, see G. Linden et. al. “Amazon.com Recommendations: Item-to-Item Collaborative Filtering” IEEE Internet Computing Industry Report, January-February 2003 (the contents of which are incorporated herein by reference). A person of ordinary skill in the art would reasonably infer that collective filtering enables a recommendation system to recommend content to a first user based on the viewing history of other users who have viewed the same media content as the first user.
Regarding the system claims 29-38 the claims are grouped and rejected with the method claims 19-28 because the steps of the method claims are met by the disclosure of the apparatus and methods of the reference(s) as discussed in the rejection of claims 19-28 and because the steps of the method are easily converted into elements of a system for providing a new content series recommendation by one of ordinary skill in the art.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ALFONSO CASTRO whose telephone number is (571)270-3950. The examiner can normally be reached on Monday to Friday from 10am to 6pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Nathan Flynn can be reached. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ALFONSO CASTRO/Primary Examiner, Art Unit 2421