Abstract: Described are embodiments for generating a post-optical proximity correction (OPC) result for a mask using a target pattern and reference layer patterns. Images of the target pattern and reference layers are provided as an input to a machine learning (ML) model to generate a post-OPC image. The images may be input separately or combined into a composite image (e.g., using a linear function) and input to the ML model. The images are rendered from pattern data. For example, a target pattern image is rendered from a target pattern to be printed on a substrate, and a reference layer image such as dummy pattern image is rendered from dummy pattern. The ML model is trained to generate the post-OPC image using multiple images associated with target patterns and reference layers, and using a reference post-OPC image of the target pattern. The post-OPC image may be used to generate a post-OPC mask.

Abstract: The present invention is directed to a method and apparatus for establishing documents relationship based on user's operation upon search result. When a user uses search engine to search for documents with a query in repository, the search result may be a list of ranked documents, and these documents may contain a lot of relationship in term of the specific query. If the user clicks some search result further, and if the click and open operation meet certain conditions, for example exceed a period of time, the clicked document could be deemed as related to the search query. Furthermore it could be inferred that there is a strong relationship between different documents clicked by the user. The present invention records the relationship between documents and presents it to the user when necessary.

Abstract: A method, system and program storage device are provided for extending an inverted index, which comprises first and second inverted index subfiles to increase the speed of establishing and updating inverted index files. The method includes performing ordered keyword indexing operations of generating an inverted index from data sources, in which a frequency of occurrence of keywords in each of the data sources is calculated, and writing each keyword, the data sources, and the frequency of occurrence of each keyword in the corresponding data sources to the inverted index. If a number of data sources involved in the indexing operations reaches a first threshold, then writing contents of the inverted index as a smallest grid into the first inverted index subfile. If a number of smallest grids in the first inverted index subfile reaches a second threshold, then merging the smallest grids into a merged grid and writing the merged grid into the second inverted index subfile.

Abstract: The present invention provides a search ranking method suitable for a file system, including receiving a query, calculating final relevance scores of individual file items with respect to the query at least partially in accordance with energy scores of individual nodes on a current file system energy tree, and outputting a list of search results based on the final relevance scores.

Abstract: The present invention is directed to a method and apparatus for establishing documents relationship based on user's operation upon search result. When a user uses search engine to search for documents with a query in repository, the search result may be a list of ranked documents, and these documents may contain a lot of relationship in term of the specific query. If the user clicks some search result further, and if the click and open operation meet certain conditions, for example exceed a period of time, the clicked document could be deemed as related to the search query. Furthermore it could be inferred that there is a strong relationship between different documents clicked by the user. The present invention records the relationship between documents and presents it to the user when necessary.

Abstract: The present invention provides a search ranking method suitable for a file system, comprising: receiving a query; calculating final relevance scores of individual file items with respect to the query at least partially in accordance with energy scores of individual nodes on a current file system energy tree, and outputting a list of search results based on the final relevance scores; and updating the file system energy tree in response to an operation on the file system performed by a user, wherein the file system energy tree has a tree structure corresponding to that of the file system, and the individual nodes thereof respectively corresponds to the individual file items in the file system. The present invention also provides a corresponding file system search engine and computer program product. With the present invention, files and file folders that the user is interested in are usually arranged in relatively higher positions of the list of search results in file system search.