A method of organizing document searches in an application's unstructured database

Abstract

The proposed technical solution pertains to the field of using large application databases (hereinafter referred to as databases) containing unstructured or weakly structured data. The proposed data structure allows for document searching using multiple keyword sets, down to the page level, without retrieving documents from the database, and presents search results in the form of an answer table, thus simplifying the process of filtering documents based on specified criteria. In this method, the program generates auxiliary tables: a keyword table, a document table, a binary string table of document numbers, and a binary string table with a supplementary inverted index. After selecting the keyword list from the inverted index, the document list presented in binary string form is processed within a module that processes the result list. Simultaneously, the binary strings are loaded into dual memory for processing, where the string's value (memory register) equals 2^n, where n is the number of bits in the memory register. If all specified keywords appear in the document, the document number equals the line number (register) in the dual memory. In the keyword module that processes document pages, an answer table is generated for each document. Dual memory is a logical memory structure containing a data input channel, an input SM, a circuit switching from ordinary memory mode to dual memory mode, an addressing module for bit numbers in memory cells, switches, and logic switches. The outputs CS, RD, and OE of the module controlling write / read are connected to the output of the bit number selection module (column address).

Claims

1. A method for organizing document searches in an application's unstructured database, comprising the following steps: The user generates a query, specifies keywords and logical operators, extracts the keywords from the query into a keyword processing block, and uses an inverted index with an index building, maintenance, and expansion program that interacts with the index block, which in turn interacts with the database. The method also includes a program that selects a list of keywords. The key features of this method are: the program generates auxiliary tables—a keyword table, a document table, binary string document numbers, and binary strings—to supplement the inverted index. The keyword table contains a list of keywords, each keyword having a binary string pointing to a document and a reference to the document table. Additionally, it specifies other information: the text of the keyword, its abbreviation, and whether it is a single or multiple element. The document table contains a list of key-value pairs associated with each keyword, where each key-value pair includes the document number using the keyword and a reference to the binary string table. The binary string table contains a fixed-length list of binary strings, each string associated with a document number in the document table. Each bit of each binary string corresponds to a page of text in the document, the bit number in the string corresponds to the page number in that document, and a 1 or 0 in that bit indicates the page number. The document's binary string table consists of multiple fixed-length binary strings, where each bit corresponds to a document number, and a 1 in that bit indicates that the document contains the given keyword. After selecting the keyword list from the inverted index, the list of documents represented as binary strings is processed in the processing results list block. During this process, the binary strings are loaded into dual memory for processing, where the string's value is equal to 2^n, where n is the number of bits in the memory register. If all specified keywords appear in documents with the same number as the string line number in memory, no list cross operation is required. In the block for processing document pages by keyword, an answer table is generated for each document. During this process, the binary strings are loaded from the binary string table into dual memory, then sorted and analyzed. In the answer table, each column corresponds to a keyword, and the columns are sorted by their importance. Each keyword column corresponds to a binary string in the binary string table. The answer table allows reflecting the keywords contained in the document at page precision without loading the document itself from the database.

2. A two-layer memory architecture for a method of organizing document search in an application's unstructured database according to claim 1, characterized in that, It is a logical memory structure that allows data to be written according to memory cell addresses, and simultaneously allows the written data to be read and output to the output line according to the memory cell addresses. Based on control signals CS, RD, OE, and SM, its characteristics include: a data input channel that allows a bit string to be written simultaneously to a specified bit in all memory cells. The specified bit refers to the column number of the bit dynamically selected by the column address signal, i.e., the position where all memory cells have the same column number; during this process, the length of the bit string is equal to the number of memory cells, the number of memory cells and the number of bits being limited by the manufacturing process of microelectronics technology; the SM input has a switching circuit function, which is in normal mode during system switching and switches to a higher mode when faster search is needed. The dual memory mode allows the circuit to switch its operating mode from ordinary memory mode to dual memory mode, writing input data to a specified bit in each memory cell. It includes an addressing block for addressing a specified bit number within a memory cell, specifying a common bit number for writing data to the specified bit in all memory cells. A switch is placed after the address block to allow or disable the memory cell address signal based on the SM control signal. It also includes logic switches installed at the inputs of all memory flip-flops to switch from the memory cell address to the address of the specified bit within the memory cell based on the SM control signal. Finally, it includes logic switches installed at the data inputs of all memory flip-flops to switch from the input data channel to the input channel of the specified bit based on the SM control signal. The outputs CS, RD, and OE of the write / read control block are connected to the outputs of the bit select block.

Images