Abstract: Applications of a domain specific model are described. A domain specific model may encode information about a domain. Information available in the domain specific model may be used to identify a topic of a conversation, such as a topic of a call to a call center. Callers' complaints can be categorized into coarse as well as fine topic categories by analyzing an initial part of a call and by examining a distribution of topic specific descriptive and discriminative features within the initial portion of the call. Once a call has been identified as belonging to a topic, a call-center agent may be prompted with information about the topic, such as questions and answers and actions related to the topic. Generic to specific information may be provided to the agent as the call progresses.

Abstract: The present disclosure provides a method for incenting potential contributors for creating content in response to a posting. The method comprises: posting a task to a first crowdsource with the task having a first expiry period of ?1; waiting for ?1 period to expire; determining whether the task is complete; reposting the task if not complete including a second expiry period of ?2; waiting for the second period of ?1 to expire; reposting the task if not yet complete including an increased reward and a third expiry period of ?3; waiting for the third period of ?3 to expire; and, reposting the task if still not complete, wherein the reposting includes a second crowdsource.

Abstract: Sentence boundaries in noisy conversational transcription data are automatically identified. Noise and transcription symbols are removed, and a training set is formed with sentence boundaries marked based on long silences or on manual markings in the transcribed data. Frequencies of head and tail n-grams that occur at the beginning and ending of sentences are determined from the training set. N-grams that occur a significant number of times in the middle of sentences in relation to their occurrences at the beginning or ending of sentences are filtered out. A boundary is marked before every head n-gram and after every tail n-gram occurring in the conversational data and remaining after filtering. Turns are identified. A boundary is marked after each turn, unless the turn ends with an impermissible tail word or is an incomplete turn. The marked boundaries in the conversational data identify sentence boundaries.

Abstract: The application discloses systems and methods for physically sharing a hard copy of a document. The systems and methods include presenting to a user a graphical user interface having printing options for printing the document, where the graphical user interface has an input for receiving an indication by the user that the user is willing to share the hard copy of the document; presenting to the user options for defining characteristics of the hard copy of the document in response to receiving the indication; and publishing at least one of the defined characteristics within a profile page of the user.

Abstract: Partial access to electronic documents and aggregation for secure document distribution is disclosed. The embodiments herein relate to providing access to electronic documents and, more particularly, to providing access to portions of electronic documents and aggregating such portions in secure document distribution environment. Existing document distribution mechanisms do not provide means to access partial documents based on the attributes such as roles of the agents within an organization, location of access, time of access, device ID and so on. The disclosed method allows agents to access partial contents of documents based on the attributes. Meta data tags are attached to the documents in order to control the access of the documents by the defined attributes.

Abstract: A keyword search system including a text input unit for inputting subtexts obtained by dividing each text into parts, while associating the subtexts with an event through a process recorded in the text; a prediction device adjuster for adjusting a corresponding event prediction device to maximize the percentage of text in which the inputted event is identical to a prediction result in a first text group selected from the subtexts; a prediction processor for generating a prediction result for each section, by inputting each text in a second text group selected from the corresponding subtexts in the adjusted event prediction device; and a search unit for calculating the prediction precision for the second text group of the event prediction device using a comparison between the inputted event and the prediction result for each subtext, and searching for keywords in sections with a certain degree of prediction precision.

Abstract: A method is provided for forming discrete segment clusters of one or more sequential sentences from a corpus of communication transcripts of transactional communications that comprises dividing the communication transcripts of the corpus into a first set of sentences spoken by a caller and a second set of sentences spoken by a responder; generating a set of sentence clusters by grouping the first and second sets of sentences according to a measure of lexical similarity using an unsupervised partitional clustering method; generating a collection of sequences of sentence types by assigning a distinct sentence type to each sentence cluster and representing each sentence of each communication transcript of the corpus with the sentence type assigned to the sentence cluster into which the sentence is grouped; and generating a specified number of discrete segment clusters by successively merging sentence clusters according to a proximity-based measure between the sentence types assigned to the sentence clusters with

Abstract: Opinions about a topic are extracted from a corpus of text documents. Opinions are extracted based on rules defining regular expressions for parts-of-speech tags. Opinions are grouped based on their semantic orientation as favorable, unfavorable or neutral. A balanced and accurate assessment of sentiment towards a topic can thus be determined.

Abstract: The embodiments of the invention provide a method for the automatic identification of changing subtopics within topics. The method begins by receiving customer satisfaction data having unstructured data objects. Next, the data objects are automatically categorized into pre-defined topics, wherein the pre-defined topics do not change throughout the customer satisfaction analysis. The pre-defined topics can be automatically defined based on a history of customer satisfaction data. Following this, a clustering analysis is automatically performed to identify subtopics of the data objects within the pre-defined topics. The subtopics are more specific than the pre-defined topics, and the subtopics can change. Further, the clustering analysis can include extracting features from the data objects and grouping the features into the subtopics. Each of the subtopics includes features having a predetermined degree of similarity.

Abstract: A method is provided for forming discrete segment clusters of one or more sequential sentences from a corpus of communication transcripts of transactional communications that comprises dividing the communication transcripts of the corpus into a first set of sentences spoken by a caller and a second set of sentences spoken by a responder; generating a set of sentence clusters by grouping the first and second sets of sentences according to a measure of lexical similarity using an unsupervised partitional clustering method; generating a collection of sequences of sentence types by assigning a distinct sentence type to each sentence cluster and representing each sentence of each communication transcript of the corpus with the sentence type assigned to the sentence cluster into which the sentence is grouped; and generating a specified number of discrete segment clusters by successively merging sentence clusters according to a proximity-based measure between the sentence types assigned to the sentence clusters with

Abstract: A method is provided for forming discrete segment clusters of one or more sequential sentences from a corpus of communication transcripts of transactional communications that comprises dividing the communication transcripts of the corpus into a first set of sentences spoken by a caller and a second set of sentences spoken by a responder; generating a specified number of sentence clusters by grouping the first and second sets of sentences according to a measure of lexical similarity using an unsupervised partitional clustering method; generating a collection of sequences of sentence types by assigning a distinct sentence type to each sentence cluster and representing each sentence of each communication transcript of the corpus with the sentence type assigned to the sentence cluster into which the sentence is grouped; and generating a specified number of discrete segment clusters by successively merging sentence clusters according to a proximity-based measure between the sentence types assigned to the sentence

Abstract: Methods, apparatus and computer programs are provided for characterizing Web-based information resources based on their interactions. A Web-based information resource is a single Web document or a collection of related Web documents. Unlike simple text documents, Web documents contain hyperlinks and other HTML tags. Different types of interactions, including inbound hyperlinks, outbound hyperlinks and internal links associated with a Web-based information resource, are used to characterize the Web-based information resource. A DOM tree representing the tag structure of a Web-based information resource is used to identify text items likely to be useful as context for a hyperlink anchor text, and the anchor text is combined with the context to generate a representation. The representation of Web-based information resources based on interactions can be used for clustering and classification, and in Web mining applications such as query disambiguation and automatic taxonomy generation.

Abstract: Sentence boundaries in noisy conversational transcription data are automatically identified. Noise and transcription symbols are removed, and a training set is formed with sentence boundaries marked based on long silences or on manual markings in the transcribed data. Frequencies of head and tail n-grams that occur at the beginning and ending of sentences are determined from the training set. N-grams that occur a significant number of times in the middle of sentences in relation to their occurrences at the beginning or ending of sentences are filtered out. A boundary is marked before every head n-gram and after every tail n-gram occurring in the conversational data and remaining after filtering. Turns are identified. A boundary is marked after each turn, unless the turn ends with an impermissible tail word or is an incomplete turn. The marked boundaries in the conversational data identify sentence boundaries.

Abstract: A keyword search system including a text input unit for inputting subtexts obtained by dividing each text into parts, while associating the subtexts with an event through a process recorded in the text; a prediction device adjuster for adjusting a corresponding event prediction device to maximize the percentage of text in which the inputted event is identical to a prediction result in a first text group selected from the subtexts; a prediction processor for generating a prediction result for each section, by inputting each text in a second text group selected from the corresponding subtexts in the adjusted event prediction device; and a search unit for calculating the prediction precision for the second text group of the event prediction device using a comparison between the inputted event and the prediction result for each subtext, and searching for keywords in sections with a certain degree of prediction precision.

Abstract: A method of building a domain specific model from transcriptions is disclosed. The method starts by applying text clustering to the transcriptions to form text clusters. The text clustering is applied at a plurality of different granularities, and groups topically similar phrases in the transcriptions. The relationship between text clusters resulting from the text clustering at different granularities is then identified to form a taxonomy. The taxonomy is augmented with topic specific information.

Abstract: A method of building a domain specific model from transcriptions is disclosed. The method starts by applying text clustering to the transcriptions to form text clusters. The text clustering is applied at a plurality of different granularities, and groups topically similar phrases in the transcriptions. The relationship between text clusters resulting from the text clustering at different granularities is then identified to form a taxonomy. The taxonomy is augmented with topic specific information.

Abstract: Web pages are previewed without actually having to browse to those web pages. A method is performed in relation to a first web page being browsed by a user and that has a hyperlink to a second web page. The second web page is acquired, and a site-specific preview, a user-specific preview, and a time-specific preview of the second web page are constructed. The site-specific preview is specific to a web site encompassing the second web page. The user-specific preview is specific to the user browsing the first web page. The time-specific preview is nominally specific to a time at which the user previews the second web page. These three previews are combined into an overall preview. In response to the user performing an action in relation to the hyperlink on the first web page, the overall preview of the second web page is displayed without browsing to that page.

Abstract: The present invention allows a user (e.g., a policy implementer) to be identified and delegated responsibility for implementing a policy. This can occur, implicitly, semi-implicitly or explicitly. In a typical embodiment, a policy provided (e.g., by a policy owner) is automatically parsed to determine a minimum set of access rights needed to implement the policy. For example, the policy might indicate that an implementing user only needs simple read privileges. Alternatively, the policy might require read/write privileges. In any event, a list (e.g., an access control list) will be analyzed to identify a set (e.g., one or more) of users of a computerized resource subject to the policy that meets the minimum set of access rights. Once this set of users has been identified, a hierarchy can be optionally analyzed to determine who among the set of users is permitted to implement the policy.

Abstract: Methods, apparatus and computer programs are provided for characterizing Web-based information resources based on their interactions. A Web-based information resource is a single Web document or a collection of related Web documents. Unlike simple text documents, Web documents contain hyperlinks and other HTML tags. Different types of interactions, including inbound hyperlinks, outbound hyperlinks and internal links associated with a Web-based information resource, are used to characterize the Web-based information resource. A DOM tree representing the tag structure of a Web-based information resource is used to identify text items likely to be useful as context for a hyperlink anchor text, and the anchor text is combined with the context to generate a representation. The representation of Web-based information resources based on interactions can be used for clustering and classification, and in Web mining applications such as query disambiguation and automatic taxonomy generation.

Abstract: Opinions about a topic are extracted from a corpus of text documents. Opinions are extracted based on rules defining regular expressions for parts-of-speech tags. Opinions are grouped based on their semantic orientation as favourable, unfavourable or neutral. A balanced and accurate assessment of sentiment towards a topic can thus be determined.