Abstract: A digital magazine server generates a digital magazine for user based on a received request for the digital magazine identifying one or more topics. The digital magazine server applies one or more machined trained models to obtained content items to select content items for the topic. A hierarchy of the topics included in the received request may be determined by the digital magazine server and used by the trained models to select content items. When generating the digital magazine, the digital magazine server also includes one or more editorial content items that are manually selected. The digital magazine serer may reposition one or more content items selected by the trained models to include an editorial content items.

Abstract: A digital magazine server generates a model to associate topics with content items. To generate the model, the digital magazine server selects a set of content items that have been included in one or more digital magazines. For each content item of the set, the digital magazine server determines a distribution of concepts associated with a content item of the set based on characteristics of digital magazines including the content item of the set and a distribution of topics associated with the content item of the set based on words included in the content item. Additionally, the digital magazine server determines a parameter defining a relationship between the distributions of concepts and the distributions of topics associated with content items of the set. A model based on the distributions of concepts and of topics as well as the parameter is generated and stored for application to content items.

Abstract: A user interaction data storage system stores user interaction data received from online systems as records. Each record of user interaction data describes user interactions performed by a user with an online system. A system, for example, a content provider system performs data access requests to access user interaction data so as to determine whether a user is likely to be interested in a content item. A plurality of online systems store a block chain comprising blocks that records transactions. Some transactions of the block chain are associated with data contribution operations and some transactions of the block chain are associated with data consumption operations.

Abstract: A digital magazine server displays content items from various sources to users of client devices. Each source of a content item is identified by a domain, and content items for different sources have different domain-level quality. To differentiate sources of content items, the domains identifying the sources are ranked based on domain scores of the domains generated by an aggregate of multiple trained domain classifiers. A domain score of a domain indicates a domain-level quality of content items provided by a source identified by the domain. Each of the trained domain classifiers (e.g., a naïve Bayes classifier, a random forest classifier, and a logistic regression classifier) generates a prediction of whether a domain is a spam domain based on the domain features and domains with known labels. Based on the domain scores of domains, the domain ranking module may adaptively select content items from the sources with corresponding domains scores.

Abstract: A digital magazine server generates a digital magazine for user based on a received request for the digital magazine identifying one or more topics. The digital magazine server applies one or more machined trained models to obtained content items to select content items for the topic. A hierarchy of the topics included in the received request may be determined by the digital magazine server and used by the trained models to select content items. When generating the digital magazine, the digital magazine server also includes one or more editorial content items that are manually selected. The digital magazine server may reposition one or more content items selected by the trained models to include an editorial content items.

Abstract: A digital magazine server separates an advertisement received for presentation to one or more digital magazine server users into a plurality of layers, each including one or more elements of the advertisement. For example, one or more foreground layers and one or more background layers of an advertisement are identified. By modifying elements included in different layers, modified advertisements associated with different display sizes are generated and associated with the display sizes. When presenting the advertisement to a user, the digital magazine server determines a size of a slot in which the advertisement is to be presented and retrieves a modified advertisement associated with the size of the slot for presentation to the user.

Abstract: A client device selects content items associated with various applications executing on the client device and dynamically lays out the selected content items for presentation while the client device is in a locked state. To present the selected content items, the client device selects a page template including slots specifying sizing and relative positioning of content items based on user attributes, geographic information, client device characteristics, and content item attributes. Content items associated with various applications are associated with each slot in the selected page template, and when the client device is in a locked state, the content items are presented with relative sizing and positioning determined by the slots in the selected page template.

Abstract: A content delivery system ensures viewability of ads in a flip-based digital magazine. User behavior in the environment of the digital magazine is observed and processed to identify one or more opportunities for ad placement. Ads are selected and placed in the digital magazine in such a way as to satisfy temporal and spatial criteria relating to display of the ads. In one embodiment, a full-page ad is inserted between consecutive content pages of the digital magazine in response to a flip action being performed by the user. The digital magazine may lock the ad in place and allow page transition after minimum display duration is satisfied. In another embodiment, an embedded ad is persisted between consecutive content pages to ensure minimum display duration is satisfied.

Abstract: A digital magazine server receives user interactions with content items presented to users via various digital magazines. The digital magazine server retrieves interactions associated with a content item that are also associated with multiple digital magazines. From the retrieved interactions, the digital magazine server generates information describing global interactions with a content item that are associated with multiple digital magazines. The digital magazine server may filter interactions with the content item that are associated with different digital magazines based on characteristics of the interactions and the information describing global interactions with the content item generated based on the filtered interactions.

Abstract: A digital magazine server displays content to a user based on a page template specifying sizing and positioning of content items. Display regions in the page template for presenting content items have a width determined by a width of a display area of a client device presenting the content and heights determined based on a percentage of the width of the display area. For example, heights of various display regions are determined as a product of a value and the percentage of the width of the display area. Heights of display regions may be determined so that different display regions have different heights.

Abstract: A digital magazine application associated with a digital magazine server executes on a client device and allows a user to interact with content from the digital magazine server even when the client device is unable to communicate with the digital magazine server. When the application detects loss of communication with the digital magazine server, interactions received by the application after the loss of communication has been detected are stored in a queue on the client device. When communication with the digital magazine server is reestablished, interactions from the queue are communicated to the digital magazine server. Certain interactions having less than a threshold measure of time dependency are stored in the queue in an order based on their associated timestamps, allowing the stored interactions to be later executed in an appropriate order by the digital magazine server.

Abstract: A digital magazine identifies content items for recommendation to a user by determining the address information of one or more content items. The address information of a content item identifies a source from which the content item was received and additional information describing the content item. Representations of content items are generated from the address information for each content item. Additionally, representations of content items previously presented to the user are generated from the address information for the previously presented content items. A measure of similarity between a representation of a content item and a representation of a previously presented content item is determined and used to select content items to present to the user.

Abstract: A digital magazine presents content items based on user interaction with or preference for content items determined based on how a user flips through different content items of the digital magazine. For example, the user may slow down or pause flipping, flip through content items at an inconsistent pace or change the navigational direction of the flipping, when the user is encountered with content items of interest. By analyzing how a user flips through different content items, content items that the user interacts with can be determined, and content items that the user may interact with or prefer can be determined and presented to the user.

Abstract: A digital magazine server scores digital magazines based on how related the digital magazines are to each of a set of topics in a topic vector space, which allows ranking for each topic the magazines that are most closely related to the topic. The digital magazine server generates a magazine vector for the magazine in the topic vector space by aggregating extracted topics for the magazine and compares the magazine vector to each of the topics to determine a magazine-topic relevance score that indicates the relevance of the digital magazine to each topic in the topic vector space. The digital magazine server then ranks each of the digital magazines by their magazine-topic relevance scores for a particular topic, e.g., a trending topic from a user request, and selects a top number of magazines for the trending topic for the user.

Abstract: A digital magazine application executing on a client device is configured to receive content items from a user of the client device or from a digital magazine server, generate pages in a digital magazine by processing the received content, and provide the pages to a user of the client device. The client device generates a new personalized digital magazine for a user by enabling the user to select content items stored in the user's local storage of the client device and to create stories about the selected content items. The personalized digital magazine includes the user's own content items, corresponding stories and content items recommended by the digital magazine server, which share common characteristics with the user's own content items. The client device can also include a user's own content items into a magazine currently presented to the user.

Abstract: A digital magazine presents content items based on user interaction with or preference for content items determined based on how a user flips through different content items of the digital magazine. For example, the user may slow down or pause flipping, flip through content items at an inconsistent pace or change the navigational direction of the flipping, when the user is encountered with content items of interest. By analyzing how a user flips through different content items, content items that the user interacts with can be determined, and content items that the user may interact with or prefer can be determined and presented to the user.

Abstract: A digital magazine server receives a request from a user to create a section of a digital magazine including information identifying additional users authorized to modify content included in the section. For example, the request includes information identifying specific additional users or includes attributes associated with additional users. Based on information in the request, the digital magazine server identifies the additional users authorized to modify the content included in the section and notifies the additional user of their authorization to modify content included in the section. A modification to content included in the section is received from an additional user (e.g., identifying a content item, modifying content of a content item in the section), with the section generated based at least in part on the received modification. When the section is presented to digital magazine server users, the additional user providing the modification is also identified.

Abstract: A digital magazine server displays content items from various sources to users of client devices. Each source of a content item is identified by a domain, and content items for different sources have different domain-level quality. To differentiate sources of content items, the domains identifying the sources are ranked based on domain scores of the domains generated by an aggregate of multiple trained domain classifiers. A domain score of a domain indicates a domain-level quality of content items provided by a source identified by the domain. Each of the trained domain classifiers (e.g., a naïve Bayes classifier, a random forest classifier, and a logistic regression classifier) generates a prediction of whether a domain is a spam domain based on the domain features and domains with known labels. Based on the domain scores of domains, the domain ranking module may adaptively select content items from the sources with corresponding domains scores.

Abstract: An online system generates implied links between objects distributed by the online system based on a plurality of users' behaviors with respect to the objects. For a behavior of a user with respect to an object that satisfies a link criterion, the online system identifies a connection between the user and the object. Local object graphs are generated to represent the relationships between the objects connected to each user, and the local object graphs are merged into a global object graph. An adjacency matrix is extracted from the global object graph as a representation of the strength of relationships between the objects distributed by the online system. The online system may then apply graph analysis techniques to the adjacency matrix, thereby analyzing the relationships between the objects without relying upon explicit links between the objects.

Abstract: A computer-implemented method is disclosed for determining, for comments associated with a content item, response likelihood scores indicating likelihood of a user responding to the comments, and presenting the comments to the user according to the determined response likelihood scores. A response likelihood score of the user responding to a comment is determined according to an emotion score associated with the comment and a user propensity of the user responding to comments with similar emotion scores for a topic of the content item. The comments associated with the content item are ranked based on the response likelihood scores associated with the comments. An arrangement of the comments can be determined based on the rankings of the comments, and the comments can be presented to the user according to the determined arrangement.