Abstract: Embodiments are disclosed for predicting target events occurrence for an advertisement campaign. A computing device according to some embodiments assigns a label to an advertisement as unlabeled, in response to a notification that a prerequisite event occurs for the advertisement. The device generates feature vectors based on data that relate to the advertisement. The device further trains a machine learning model using the feature vectors of the unlabeled advertisement based on a first term of an objective function, without waiting for a target event for the advertisement to occur. The first term depends on unlabeled advertisements. The device predicts a probability of a target event occurring for a new advertisement, by feeding data of the new advertisement to the trained machine learning model.

Abstract: Embodiments are disclosed for predicting target events occurrence for an advertisement campaign. A computing device according to some embodiments assigns a label to an advertisement as unlabeled, in response to a notification that a prerequisite event occurs for the advertisement. The device generates feature vectors based on data that relate to the advertisement. The device further trains a machine learning model using the feature vectors of the unlabeled advertisement based on a first term of an objective function, without waiting for a target event for the advertisement to occur. The first term depends on unlabeled advertisements. The device predicts a probability of a target event occurring for a new advertisement, by feeding data of the new advertisement to the trained machine learning model.

Abstract: Systems, methods, and computer-readable storage media for presenting social network connections in association with a search engine results page (SERP) are provided. Upon receipt of a search query, it is determined if the query is a name query and if social networking data associated with the user is available. If it is determined that the query is a name query and that social networking data associated with the user is available, it is determined if there is a connection match to the query. If it is determined that one or more social network connections of the user match the name query, the matching social network connection is presented in association with the SERP. Additionally, the user may add a presented social network connection as a first degree connection and/or send a message to a presented connection from the context of the SERP.

Abstract: Architecture that enables extraction of pieces of content from the Internet, social networks, in particular, and/or of a user's immediate circle of friends and family, where the content relates to trending topics of those networks as relevant to a search query. The content effectively summarizes a trending topic (e.g., of a single social network or across multiple social networks). The summary can be characterized by different content types, such as images only, text only, and so on. The extracted content can then be presented in a number of different ways such as in a display that visually rotates through the different types of content, such as trending images, summarizing updates, and trending webpages, for example. Another technique for presentation of the extracted social content can be as user-selectable “hotspots” on a page or desktop, and which can display the trending topic summaries.

Abstract: Systems, computer-readable media, and methods for tagging documents based on a graph pertaining to one or more entities which a user has included in a search query. The user may have at least one social networking relationship with the entity. A search engine is configured to display a search engine results page in response to the search query received from the user. The search engine may also receive suggested tags that identify documents that could be linked to the entity identified in the query. The user may confirm that the suggested tags are appropriate via feedback that is transmitted to the search engine. In turn, the search engine updates a graph to reflect a number of users that agree with the suggested tag.

Abstract: Methods, systems, and computer-storage media having computer-usable instructions embodied thereon, for prioritizing crawl lists based on social networking rankings are provided. Various scores are associated with users based on a variety of factors including activity levels with respect to social networking services, activity levels with respect to search engines, and interactions with other users in a social networking environment. The scores are used to compute a ranking for the users and, based on the rankings, a crawl list is prioritized such that content associated with the social networking environment is crawled at an appropriate time.

Abstract: Systems, methods, and computer-readable storage media for tagging uniform resource locators (URLs) included in a search engine results page are provided. Upon receipt of a search query that includes a name of an entity, the server device provides search results and entry fields to tag one or more URLs that correspond to the entity. The server device obtains a social network of the user that provided the query. The server device processes the social network to locate entities that match the query and that are within a threshold number of hops. The user may tag one or more URLs with social network account information for the entities based on authorization information received from an owner that manages the social network account.

Abstract: Systems, methods, and computer-readable storage media for presenting social network connections in association with a search engine results page (SERP) are provided. Upon receipt of a search query, it is determined if the query is a name query and if social networking data associated with the user is available. If it is determined that the query is a name query and that social networking data associated with the user is available, it is determined if there is a connection match to the query. If it is determined that one or more social network connections of the user match the name query, the matching social network connection is presented in association with the SERP. Additionally, the user may add a presented social network connection as a first degree connection and/or send a message to a presented connection from the context of the SERP.

Abstract: Methods, computer systems and computer-readable media for adding social network data to search suggestions are provided. At least a portion of a search query is received, and a set of suggested search queries is determined for the at least a portion of the search query. A first suggested search query in the set of suggested search queries is determined to be associated with a person or entity. A social data store is accessed; the social data store includes social network data. It is determined that the person or entity is associated with a set of social network data in the social data store, and the set of social network data in association with the first suggested search query is provided to a search engine page.

Abstract: Links to web content can be identified as a function of one or more links shared by a user of an online social network service, among other things. The identified links can represent recommended links likely to be interesting to the user. Densification techniques can be employed to address data sparsity and thus enhance link recommendation. Furthermore, recommended links can be integrated with a search engine to personalize interaction with web content.

Abstract: Methods, systems, and computer-storage media having computer-usable instructions embodied thereon, for prioritizing crawl lists based on social networking rankings are provided. Various scores are associated with users based on a variety of factors including activity levels with respect to social networking services, activity levels with respect to search engines, and interactions with other users in a social networking environment. The scores are used to compute a ranking for the users and, based on the rankings, a crawl list is prioritized such that content associated with the social networking environment is crawled at an appropriate time.