Abstract: Improved solutions for dataset clustering and evaluation are disclosed. Examples cluster a set of documents into set of clusters using a language model, in an iterative process. In second and later clustering tasks, the current cluster titles and descriptions are provided in the language model prompt, to avoid near-duplications. Upon determining that the set of clusters is sufficiently complete and representative of the set of documents, the tasking switches to classification of the set of documents into the set of clusters using a language model. Classification continues until a sufficient percentage of the set of documents is classified. Some examples use batching, to avoid overloading the language model(s). In some examples, different language models are used for clustering and classification. Some examples use intruder detection to determine the quality of the clustering.

Abstract: Improved solutions for dataset clustering and evaluation are disclosed. Examples cluster a set of documents into set of clusters using a language model, in an iterative process. In second and later clustering tasks, the current cluster titles and descriptions are provided in the language model prompt, to avoid near-duplications. Upon determining that the set of clusters is sufficiently complete and representative of the set of documents, the tasking switches to classification of the set of documents into the set of clusters using a language model. Classification continues until a sufficient percentage of the set of documents is classified. Some examples use batching, to avoid overloading the language model(s). In some examples, different language models are used for clustering and classification. Some examples use intruder detection to determine the quality of the clustering.

Abstract: According to an example aspect of the present invention, there is provided an apparatus comprising at least one processor; and at least one memory including computer program code; the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to function as a point of interception in an application server or border control function of a communication network, receive an incoming protocol message requesting initiation of a call, transmit an outgoing protocol message to advance initiation of the call, and receive a cryptographic token comprising a cryptographically signed identity of a caller initiating the call, and transmit a lawful interception message comprising information on the call to a lawful interception party as a response to at least one trigger being fulfilled.

Abstract: Systems and methods are directed to building annotated models based on eyes-off data. Specifically, a synthetic data generation model is trained and used to further train a target model. The synthetic data generation model is trained within an eyes-off environment using an anonymity technique on confidential data. The synthetic data generation model is then used to create synthetic data that closely represents the confidential data but without any specific details that can be linked back to the confidential data. The synthetic data is then annotated and used to train the target model within an eyes-on environment. Subsequently, the target model is deployed back within the eyes-off environment to classify the confidential data.

Abstract: A method and system for generating synthetic privacy preserving training data for training a language classifier machine-learning (ML) model includes receiving a request to generate the synthetic privacy-preserving training data for the language classifier ML model, retrieving labeled training data associated with training the language classifier ML model, providing the labeled training data, one or more privacy parameters, and a domain type associated with the labeled training data to a synthetic data generation ML model, the synthetic data generation ML model being configured to generate synthetic training data in a privacy-persevering manner, receiving synthetic privacy-preserving training data as an output from the synthetic data generation ML model, and providing the synthetic privacy preserving training data to the language classifier ML model for training the language classifier ML model in classifying text.

Abstract: Systems and methods are directed to building annotated models based on eyes-off data. Specifically, a synthetic data generation model is trained and used to further train a target model. The synthetic data generation model is trained within an eyes-off environment using an anonymity technique on confidential data. The synthetic data generation model is then used to create synthetic data that closely represents the confidential data but without any specific details that can be linked back to the confidential data. The synthetic data is then annotated and used to train the target model within an eyes-on environment. Subsequently, the target model is deployed back within the eyes-off environment to classify the confidential data.

Abstract: Systems and methods are directed to building annotated models based on eyes-off data. Specifically, a synthetic data generation model is trained and used to further train a target model. The synthetic data generation model is trained within an eyes-off environment using an anonymity technique on confidential data. The synthetic data generation model is then used to create synthetic data that closely represents the confidential data but without any specific details that can be linked back to the confidential data. The synthetic data is then annotated and used to train the target model within an eyes-on environment. Subsequently, the target model is deployed back within the eyes-off environment to classify the confidential data.

Abstract: According to an example aspect of the present invention, there is provided an apparatus comprising at least one processor; and at least one memory including computer program code; the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to function as a point of interception in an application server or border control function of a communication network, receive an incoming protocol message requesting initiation of a call, transmit an outgoing protocol message to advance initiation of the call, and receive a cryptographic token comprising a cryptographically signed identity of a caller initiating the call, and transmit a lawful interception message comprising information on the call to a lawful interception party as a response to at least one trigger being fulfilled.

Abstract: A method and system for generating synthetic privacy preserving training data for training a language classifier machine-learning (ML) model includes receiving a request to generate the synthetic privacy-preserving training data for the language classifier ML model, retrieving labeled training data associated with training the language classifier ML model, providing the labeled training data, one or more privacy parameters, and a domain type associated with the labeled training data to a synthetic data generation ML model, the synthetic data generation ML model being configured to generate synthetic training data in a privacy-persevering manner, receiving synthetic privacy-preserving training data as an output from the synthetic data generation ML model, and providing the synthetic privacy preserving training data to the language classifier ML model for training the language classifier ML model in classifying text.

Abstract: Disclosed is a method and a system for extracting entity information from target data. The method comprises: providing the target data; refining the target data to obtain at least one base entity information having a plurality of base entity units using an algorithm, wherein the algorithm is based on a predefined syntax; generating a plurality of strings for each of the base entity information, wherein the plurality of strings comprises at least one base entity unit among the plurality of base entity units; sorting the plurality of strings in a decreasing order of length of the plurality of strings; identifying an entity type of the plurality of strings, based on an ontology, by processing the plurality of strings sequentially; assigning labels to the plurality of strings based on the entity type; and mapping the labelled plurality of strings to a predefined signature to obtain the entity information.

Abstract: In one embodiment, a method includes receiving a search query input comprising one or more n-grams; parsing the search query input to identify keywords; generating query commands for the keywords. Each query command may specify: a particular object-type; one or more identifiers of one or more objects that match the search query input; and one or more types of relationships with respect to the objects. The method may further include searching a particular vertical that stores objects of the particular object-type having a relationship of the type of relationship with respect to one or more of the objects; generating a plurality of search-result modules corresponding to the query commands, each search-result module comprising references to objects of the particular object-type specified by the query command; and sending, to a client device, instructions for presenting an interface comprising one or more of the search-result modules.

Abstract: Disclosed is a method and a system for extracting entity information from target data. The method comprises: providing the target data; refining the target data to obtain at least one base entity information having a plurality of base entity units using an algorithm, wherein the algorithm is based on a predefined syntax; generating a plurality of strings for each of the base entity information, wherein the plurality of strings comprises at least one base entity unit among the plurality of base entity units; sorting the plurality of strings in a decreasing order of length of the plurality of strings; identifying an entity type of the plurality of strings, based on an ontology, by processing the plurality of strings sequentially; assigning labels to the plurality of strings based on the entity type; and mapping the labelled plurality of strings to a predefined signature to obtain the entity information.

Abstract: Exemplary embodiments relate to techniques that allow for file system support to be rapidly deployed for new or updated operating system distributions. In some embodiments, a management component is provided perform data management on file systems. When a data management operation on a file system is requested, an operation component searches in a predetermined location for a named module that implements certain types of operations. The operation component then calls these operations (including validate, build and deport operations for the file system) to implement data management procedures in the file system. Implementing support for a new operating system or file system does not require that the management entity be rebuilt. Upon release of a new operating system or file system, a new named module can be written and placed in the predetermined location where the operation module is configured to search.

Abstract: Exemplary embodiments relate to techniques that allow for file system support to be rapidly deployed for new or updated operating system distributions. In some embodiments, a management component is provided perform data management on file systems. When a data management operation on a file system is requested, an operation component searches in a predetermined location for a named module that implements certain types of operations. The operation component then calls these operations (including validate, build and deport operations for the file system) to implement data management procedures in the file system. Implementing support for a new operating system or file system does not require that the management entity be rebuilt. Upon release of a new operating system or file system, a new named module can be written and placed in the predetermined location where the operation module is configured to search.

Abstract: In one embodiment, a method includes receiving a search query input comprising one or more n-grams; parsing the search query input to identify keywords; generating query commands for the keywords. Each query command may specify: a particular object-type; one or more identifiers of one or more objects that match the search query input; and one or more types of relationships with respect to the objects. The method may further include searching a particular vertical that stores objects of the particular object-type having a relationship of the type of relationship with respect to one or more of the objects; generating a plurality of search-result modules corresponding to the query commands, each search-result module comprising references to objects of the particular object-type specified by the query command; and sending, to a client device, instructions for presenting an interface comprising one or more of the search-result modules.

Abstract: In one embodiment, a method includes receiving from a first user of an online social network a search query input including one or more n-grams; generating a number of query commands based on the search query input; and searching one or more verticals to identify one or more objects stored by the vertical that match the query commands. Each vertical stores one or more objects associated with the online social network. The method also includes generating a number of search-result modules. Each search-result module corresponds to a query command of the number of query commands. Each search-result module includes references to one or more of the identified objects matching the query command corresponding to the search-result module. The method also includes scoring the search-result modules; and sending each search-result module having a score greater than a threshold score to the first user for display.

Abstract: Computer-readable media, computer systems, and computing methods are provided for classifying search results as either of good quality or of poor quality. Initially, a portion of the search results, such as the highest ranked documents, are selected for evaluation. A level of quality for each of the selected search results is determined using a classification process that includes the following steps: targeting features demonstrated by the selected search results to be evaluated; evaluating the selected features to generate a level-of-quality score for each of the selected search results; comparing the score against a predefined threshold value; and, based on the comparison, assigning each of the selected search results an absolute measurement. The absolute measurement indicates poor quality when the score is less than the threshold value. Upon recognizing that the selected search results are of poor quality, automatically executing a corrective action that reformulates the issued search query.

Abstract: In one embodiment, a method includes receiving from a first user of an online social network a search query input including one or more n-grams; generating a number of query commands based on the search query input; and searching one or more verticals to identify one or more objects stored by the vertical that match the query commands. Each vertical stores one or more objects associated with the online social network. The method also includes generating a number of search-result modules. Each search-result module corresponds to a query command of the number of query commands. Each search-result module includes references to one or more of the identified objects matching the query command corresponding to the search-result module. The method also includes scoring the search-result modules; and sending each search-result module having a score greater than a threshold score to the first user for display.

Abstract: Subject matter described herein is related to determining a document score, which suggests a relevance of a document (e.g., webpage) to a search query. For example, a search query is received that is comprised of one or more terms, which represent a subject. An equivalent subject is identified that is semantically similar to the subject. The document score is determined by accounting for both a subject frequency and an equivalent-subject frequency.

Abstract: Techniques are disclosed for providing a domain-aware snippet for a search result. A uniform resource locator (URL) is identified for a search result obtained in response to a search query, and it is determined that the URL corresponds to a single domain that has a plurality of web pages that are generated using a template that is common to each of the web pages in the domain. The template comprises a hypertext markup language (HTML) layout pattern that includes multiple sections shared by the web pages. A ranking value is assigned to the multiple sections and is used to identify a first section of the template that is relevant to the search query. A snippet is provided to a user for the search result; the snippet includes at least a portion of text from the first section.