摘要:
Controlling cascade of information transmitted and received via one or more online data sharing platforms and electronic communication network. A message to be shared is detected. The message and the corresponding message stream may be analyzed. An action may be taken upon the message responsive to one or more criteria being met based on analyzing the message and the corresponding message stream. The action may include at least forcing a fact check of the message that determines veracity of the message. The action may also include one or more of removing the message, disabling the message, allowing conditional sharing of the message, or allowing the message to be shared. The one or more criteria may include at least determining a confidence value below a threshold value that indicates the message is non-factual.
摘要:
For managing a pool of virtual computer systems, master status and a rebasing task are assigned to at least one virtual computer system of the pool. A non-reserved status is assigned to virtual computer systems of the pool that are not assigned tasks. A virtual computer system of the pool is selected to respond to a rebasing request. The selecting includes selecting one of the at least one virtual computer system that currently has a master status and that currently has a processing load less than a predetermined threshold and, when no virtual computer system of the pool currently has a master status and a processing load less than the predetermined threshold, selecting a non-reserved one of the virtual computer systems to perform rebasing for the request and changing the non-reserved status of the selected, non-reserved one of the virtual computer systems to master status.
摘要:
Electronic message text classification framework selection is described. An incoming electronic message is classified using a current text classification framework. A classification of the electronic message by the current text classification framework is scored. A cost of re-training the current text classification is compared against a cost of switching to a different text classification framework. One of multiple text classification frameworks, which includes the current text classification framework and other text classification frameworks, is selected based on the score of the classification by the current text classification framework and a result of the comparison.
摘要:
Electronic message text classification framework selection is described. An incoming electronic message is classified using a current text classification framework. A classification of the electronic message by the current text classification framework is scored. A cost of re-training the current text classification is compared against a cost of switching to a different text classification framework. One of multiple text classification frameworks, which includes the current text classification framework and other text classification frameworks, is selected based on the score of the classification by the current text classification framework and a result of the comparison.
摘要:
A method for generating a graph segment providing a gist or summary of an online social network conversation may include generating a graph of the online social network conversation. The graph of the online social network conversation may include a plurality of nodes and each node may be connected to at least one other node by an edge. Each node may represent a message of the online social network conversation and each edge may correspond to an action by a participant in the online social network conversation. The method may also include determining an edge weight for each edge and analyzing the graph of the online social network conversation using at least the edge weight of at least some edges. The method may additionally include generating a graph segment that provides a gist or summary of the online social network conversation based on the analysis.
摘要:
Controlling cascade of information transmitted and received via one or more online data sharing platforms and electronic communication network. A message to be shared is detected. The message and the corresponding message stream may be analyzed. An action may be taken upon the message responsive to one or more criteria being met based on analyzing the message and the corresponding message stream. The action may include at least forcing a fact check of the message that determines veracity of the message. The action may also include one or more of removing the message, disabling the message, allowing conditional sharing of the message, or allowing the message to be shared. The one or more criteria may include at least determining a confidence value below a threshold value that indicates the message is non-factual.
摘要:
Mapping and obscuring digital representations of a number of user accounts on a social network map includes identifying a primary user account from a number of user accounts of a social network, determining, based on metadata associated with the user accounts, a relationship for each of the user accounts relative to the primary user account, mapping, based on the relationship for each of the user accounts relative to the primary user account, a digital representation of each of the user accounts to a territory on a social network map, determining, based on the relationship for each of the user accounts relative to the primary user account, an obscurity level for each of the user accounts, and obscuring, based on the obscurity level, the territory associated with the digital representation of each of the user accounts on a social network map from the primary user account.
摘要:
A method, system and computer program product for providing enhanced action fulfillment using classification valency. At least a first message from a user in a channel is analyzed, the first message containing first data of a first data type, and a classification of the first message is inferred based on the first data's content. At least a second message from the user in the channel is analyzed, the second message containing second data of a second data type different from the first data type, and a classification model of the second message is derived based on the second data's content. The classification of the first message and the classification model of the second message are used to determine whether an additional inference is available based on the classification valency. The additional inference is incorporated to determine the enhanced action fulfillment, and the enhanced action is presented to the user for fulfillment.
摘要:
Mapping and obscuring digital representations of a number of user accounts on a social network map includes identifying a primary user account from a number of user accounts of a social network, determining, based on metadata associated with the user accounts, a relationship for each of the user accounts relative to the primary user account, mapping, based on the relationship for each of the user accounts relative to the primary user account, a digital representation of each of the user accounts to a territory on a social network map, determining, based on the relationship for each of the user accounts relative to the primary user account, an obscurity level for each of the user accounts, and obscuring, based on the obscurity level, the territory associated with the digital representation of each of the user accounts on a social network map from the primary user account.
摘要:
A method for generating a graph segment providing a gist or summary of an online social network conversation may include generating a graph of the online social network conversation. The graph of the online social network conversation may include a plurality of nodes and each node may be connected to at least one other node by an edge. Each node may represent a message of the online social network conversation and each edge may correspond to an action by a participant in the online social network conversation. The method may also include determining an edge weight for each edge and analyzing the graph of the online social network conversation using at least the edge weight of at least some edges. The method may additionally include generating a graph segment that provides a gist or summary of the online social network conversation based on the analysis.