Abstract: A computer-implemented method determines the resemblance of data objects such as Web pages. Each data object is partitioned into a sequence of tokens. The tokens are grouped into overlapping sets of the tokens to form shingles. Each shingle is represented by a unique identification element encoded as a fingerprint. A minimum element from each of the images of the set of fingerprints associated with a document under each of a plurality of pseudo random permutations of the set of all fingerprints are selected to generate a sketch of each data object. The sketches characterize the resemblance of the data objects. The sketches can be further partitioned into a plurality of groups. Each group is fingerprinted to form a feature. Data objects that share more than a certain numbers of features are estimated to be nearly identical.

Abstract: A method and computer product for automatically generating an IP network address that facilitates simplified network connection and administration for small-scale IP networks without IP address servers, such as those found in a small business or home network environment. First, a proposed IP address is generated by selecting a network identifying portion (sometimes known as an IP network prefix) while deterministically generating the host identifying portion based on information available to the IP host. For example, the IEEE 802 Ethernet address found in the network interface card may be used with a deterministic hashing function to generate the host identifying portion of the IP address. Next, the generated IP address is tested on the network to assure that no existing IP host is using that particular IP address. If the generated IP address already exists, then a new IP address is generated, otherwise, the IP host will use the generated IP address to communicate over the network.

Abstract: A system for automatically providing hypertext for character strings of a text file at a content server. A central server provides central control of the links of text files of a plurality of content servers in an information network such as the Internet. The central server intermittently updates each content server with new character strings and/or destination addresses, such as Uniform Resource Locators (URLs). The content servers also update the central server with new character strings. Optionally, each content server can query the central server on a real-time basis to obtain a destination address for a character string which does not have a corresponding valid destination address. The central server responds to such queries by searching its master databases, and using a search engine if required. Hit count data is maintained at the content servers and transmitted to the central server intermittently.

Abstract: A method of retrieving documents from a document database is disclosed. A set of documents is retrieved according to a first search statement. A signature for a first retrieved document, and preferably other documents by searching for words in the first document and removing common words which occur in a relatively high frequency in a natural language in which the first document is written. The document for which the signature was developed is displayed. Responsive to a user indication that a second search is to be made, deriving a second search statement from the signature of the document.In the preferred embodiment, a "spectrum" of documents is prepared and presented to the user. The signatures of a plurality of documents from the documents retrieved according to the first search statement by searching for words in the documents and removing common words which occur in a relatively high frequency in a natural language in which the documents are written.

Abstract: Structured graphical data is reorganised. The data, which may be defined in accordance with portable document format (PDF) includes graphical object definitions and references to said definitions. The data is reorganised so that the graphical object references are preceded by their respective object definitions.

Abstract: A kana-to-kanji conversion system, in which an input character string is displayed in Romaji in the order in which the characters are inputted in a client on a network, the input character string in Romaji is transmitted successively from the client to a server, a kana character string is obtained by processing the successively transmitted input character string in the server, the kana character string is returned to the originator client from the server, and, of the displayed Romaji in the client, a display of a section corresponding to the returned kana character string is changed to the kana character string.

Abstract: Mechanisms and methods for storing, dynamically reconstructing, and navigating a three-dimensional virtual world using a database are disclosed. A virtual world is described in a source text according to the grammar of a modeling language. The source text is read, parsed, and decomposed into a database schema in which characteristics of the world are represented in database tables. In an embodiment, nodes and fields of the world are associated with database queries. When the world is to be displayed, values in the database schema are recomposed into a source text. The database queries are executed against a database, yielding values, in real time based on the current state of the data in the database, for the nodes associated with the queries. Thus, a large virtual world are efficiently displayed and easily modified, and the size, shape, or other aspects of elements of the virtual world can change as data in the database changes.

Abstract: A method of data mining represents related items in a multidimensional space. Distance between items in the multidimensional space corresponds to the extent of relationship between the items. The user can select portions of the space to perceive. The user also can interact with and control the communication of the space, focusing attention on aspects of the space of most interest. The multidimensional spatial representation allows more ready comprehension of the structure of the relationships among the items.

Abstract: A supra-search engine tool includes a distributed computer system and automatically structures and organizes information requests, then independently searches, requests and organizes the data from information providers, including a variety of search engines and websites to match the tailored requests of the information consumers. The retrieved and processed information is then accessible via website, browser, fax, email, voicemail, mail, software and other communication means.

Abstract: A data management system and method enables the storage of long records in a set of keyed physical records of restricted length while minimising movement of data. The logical record to be stored is logically divided into a number of physical record portions to each of which is prepended a key with a unique sequence number. By starting from one end of the record with the key of highest sequence number and copying the physical record consisting of key plus data into the data set, successive physical records can be assembled in situ by overwriting the previous record's data portion with the current record's key. This ensures that the split logical record data need only be moved once as it is transferred to non-volatile storage as physical records of the data set. The original logical record can be reassembled by reversing the above procedure.