Abstract: A keyboard utilizing a matrix arrangement of word, character, numerical (print medium) keys to enhance a standard print medium keyboard. Instead of utilizing keys with letters and numbers, and instead of utilizing iconic symbols (or for use in combination with iconic symbols), the keyboard includes a plurality of keys arranged in at least two matrices. The keys of the first matrix are associated with each other in some manner, as are keys in at least one additional matrix. The matrices can include keys grouped together and grammatically associated with one another, by part of speech for example; and/or can include keys associated with a common color and a common number of selections necessary to access an associated word or word phrase. By use of such a keyboard, preferably in conjunction with a word output device, both manual and scanning input can be enhanced in a print medium environment, without the need to invest substantial time in memorizing complex iconic structures.

Abstract: A system, method, and computer program product for determining the emotional content of an electronic correspondence to route or prioritize the information, to set the expectations of a customer support worker, to flag those workers who are using inappropriate language with the customer, or to determine another best course to send the correspondence. In a preferred embodiment, a customer sends an electronic correspondence to a company via email. Emotionally charged words or symbols in each sentence are detected. The message is then given an emotional ranking which is used to determine what future action is most appropriate for the correspondence.

Abstract: Starting with a dictionary including concepts and chains, a subset of the chains is selected as intentional stance basis chains (ISBCs). Concepts in the dictionary are chosen and mapped to state vectors using the ISBCs. The state vectors are assembled into a template, and an action and threshold distance are assigned to the template. When an impact summary is generated for a content source, the distance between the impact summary and the template is measured. If the distance is less than the threshold distance associated with the action for the template, the action is performed.

Abstract: A method of generating a phonemic transcription for a word using a computer system is described. In one embodiment, an existing pronunciation generation program is applied to generate an initial transcription. The initial transcription can then be evaluated to identify likely bad pronunciations by looking for phonotactically impossible co-occurrences. Additionally, one or more rules can be applied to generate additional phonemic transcriptions. The resulting transcriptions may be used in place of the initial transcription and/or in addition to the initial transcription. Additionally, when multiple transcriptions result, the transcriptions are ordered according to preference and/or likelihood of use.

Abstract: A method for the joint optimization of language model performance and size is presented comprising developing a language model from a tuning set of information, segmenting at least a subset of a received textual corpus and calculating a perplexity value for each segment and refining the language model with one or more segments of the received corpus based, at least in part, on the calculated perplexity value for the one or more segments.

Abstract: A computer-implemented system and method is disclosed for retrieving documents using context-dependant probabilistic modeling of words and documents. The present invention uses multiple overlapping vectors to represent each document. Each vector is centered on each of the words in the document and includes the local environment. The vectors are used to build probability models that are used for predictions of related documents and related keywords. The results of the statistical analysis are used for retrieving an indexed document, for extracting features from a document, or for finding a word within a document. The statistical evaluation is also used to evaluate the probability of relation between the key words appearing in the document and building a vocabulary of key words that are generally found together. The results of the analysis are stored in a repository. Searches of the data repository produce a list of related documents and a list of related terms.

Abstract: The present invention relates to packet-distributed data transmission of compressed data. According to the invention, parity bits are supplied to the compressed data. The parity bits are used in the entire transmission chain between an encoder having compressed the data, and a decoder which decompresses it. According to one embodiment, the data is speech and the packet-distributed network is a mobile radio network with packet-distribution in included links. However, sending in the radio link of the compressed speech is circuit switched.

Abstract: Codifying the “most prominent measurement points” of a document can be used to measure semantic distances given an area of study (e.g., white papers on some subject area). A semantic abstract is created for each document. The semantic abstract is a semantic measure of the subject or theme of the document providing a new and unique mechanism for characterizing content. The semantic abstract includes state vectors in the topological vector space, each state vector representing one lexeme or lexeme phrase about the document. The state vectors can be dominant phrase vectors in the topological vector space mapped from dominant phrases extracted from the document. The state vectors can also correspond to words in the document that are most significant to the document's meaning (the state vectors are called dominant vectors in this case). One semantic abstract can be directly compared with another semantic abstract, resulting in a numeric semantic distance between the semantic abstracts being compared.

Abstract: A system, method and article of manufacture utilize word chunks to enhance word prediction. At least one of selectable words and word chunks are displayed in response to receipt of an input character. Thereafter, a selection of a displayed word or word chunk is received, and in response to receiving selection of a displayed word chunk, selectable words including the selected word chunk are displayed. A word chunk includes a word portion used in the formation of other words and includes a predetermined identifier, identifying it as a word chuck. Thus, the combination of chunking of parts of agglutinated words in combination with letter input activations reduces the overall number of input activations to complete a word, and further reduces the amount of choices to select from at any point during retrieval of the word.

Abstract: An evaluator system accepts input textual messages in unknown languages and assesses which character sets, corresponding to languages, matches that message. Textual messages whose individual characters are encoded in 16 bit Unicode or other universal format are parsed, and character sets which can express each character and the accumulated correspondence is logged. When the character sets against which the message is being tested only provide partial matches, the invention can determine which offers the best fit, including by means of a weighting function. The evaluation technology of the invention can be applied to multipart documents, and to search engines and indices.

Abstract: A system and method for determining the language of an unknown document is provided. For a set of candidate languages, a negative assumption is set for each candidate language that the document is not that language and the system attempts to prove the negative assumption is wrong. If the negative assumption fails for one language, then the document is identified as being in that language. The present system and method provides a higher degree of accuracy when determining the language of a document.

Abstract: Word sense ambiguity, for “thematic” words in a sentence, is achieved based on thematic prediction. The senses of “thematic” words are disambiguated in a sentence by determining and weighting possible themes for that sentence. Possible themes are determined for that sentence based on thematic information associated with the different senses of each word in the sentence. A highly deterministic thematic-based word sense disambiguation method is used to preprocess the sentence prior to further syntactic and semantic analysis, thereby enhancing accuracy and decreasing the demand for computational resources (memory and CPU) by reducing input ambiguities.

Abstract: A method of analyzing an author's work, including reading a text file, generating an analysis model from the text file and storing the analysis model. The text file may be a poem, the poem containing ASCII text. The analysis model is a linked data structure. The linked data structure includes n-gram data structures, the n-gram structures including combinations of 1-gram data structures, bigram data structures, trigram data structures and quadrigram data structures.

Abstract: A system, method and article of manufacture utilize word chunks to enhance word prediction. At least one of selectable words and word chunks are displayed in response to receipt of an input character. Thereafter, a selection of a displayed word or word chunk is received, and in response to receiving selection of a displayed word chunk, selectable words including the selected word chunk are displayed. A word chunk includes a word portion used in the formation of other words and includes a predetermined identifier, identifying it as a word chuck. Thus, the combination of chunking of parts of agglutinated words in combination with letter input activations reduces the overall number of input activations to complete a word, and further reduces the amount of choices to select from at any point during retrieval of the word.

Abstract: An audio-based data indexing and retrieval system for processing audio-based data associated with a particular language, comprising: (i) memory for storing the audio-based data; (ii) a semantic unit based speech recognition system for generating a textual representation of the audio-based data, the textual representation being in the form of one or more semantic units corresponding to the audio-based data; (iii) an indexing and storage module, operatively coupled to the semantic unit based speech recognition system and the memory, for indexing the one or more semantic units and storing the one or more indexed semantic units; and (iv) a search engine, operatively coupled to the indexing and storage module and the memory, for searching the one or more indexed semantic units for a match with one or more semantic units associated with a user query, and for retrieving the stored audio based data based on the one or more indexed semantic units. The semantic unit may preferably be a syllable or morpheme.

Abstract: A Web browser obtains a document in which a pointer indicative of an address in a translating unit has been defined among documents from an HTTP server unit. When translation is instructed by the user, the Web browser transmits a translation request message together with a URL of the document to the translating unit. The translating unit obtains the document corresponding to the supplied URL and transmits a translation result to the Web browser. A document whose translation is not permitted is prevented from being translated.

Abstract: Initials of the Chinese language are classified into nine initial groups. The first touch of a numeric key of a cellular phone selects an initial group including the initial of the desired syllable for input. Then, guidance is displayed corresponding to a final group to which the final of the desired syllable belongs. The syllable is determined based on the operation according to the guidance. The present invention classifies finals into the first final group composed of simple finals, the second final group composed of complex finals that start with a, e, or o, the third final group composed of complex finals that start with i, or y, the fourth final group composed of complex finals that start with u, or w, and the fifth final group composed of complex finals that start with v, or yu.

Abstract: Initials of the Chinese language are classified into nine initial groups. The first touch of a numeric key of a cellular phone selects an initial group including the initial of the desired syllable for input. The second touch of a numeric key selects one of initial-final groups, each including an initial and one of five final groups. These final groups comprise a simple final group and four types of complex final groups. The third touch of a numeric key specifies a final to determine the desired syllable.

Abstract: A learning system has at least one printed entity comprising lesson information in an ordered series of one or more of sentences, phrases, pictures, or alphanumeric representations, a digital memory media having recorded thereon separately playable audio files comprising one or both of vocal renditions of individual ones of the sentences or phrases, vocal renditions related to entities in the pictures, or information files, and a digital player for playing the audio files from the memory media, rendering the audio files audible, and having a user interface for selecting and initiating playing of individual ones of the files, individually or in a predetermined order.

Abstract: A directed set can be used to establish contexts for linguistic concepts: for example, to aid in answering a question, to refine a query, or even to determine what questions can be answered given certain knowledge. A directed set includes a plurality of elements and chains relating the concepts. One concept is identified as a maximal element. The chains connect the maximal element to each concept in the directed set, and more than one chain can connect the maximal element to any individual concept either directly or through one or more intermediate concepts. A subset of the chains is selected to form a basis for the directed set. Each concept in the directed set is measured to determine how concretely each chain in the basis represents it. These measurements for a single concept form a vector in Euclidean k-space. Distances between these vectors can be used to determine how closely related pairs of concepts are in the directed set.