# US 10,482,162 B2 · Automatic equation transformation from text > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,482,162 B2 - Document ID: US10482162B2 - Status: Granted - Kind: B2 - Application number: 15/828,342 - Filed: 2017-11-30 - Published: 2019-05-30 - Granted: 2019-11-19 - Earliest claimed: 2017-11-30 - Assignees: International Business Machines Corporation - Applicants: International Business Machines Corporation - Inventors: Keke Cai; HongLei Guo; Zhong Su; Li Zhang; Shiwan Zhao - Web page: https://patentgrep.com/patent/US10482162B2 ## CPC - G06F 40/111: Mathematical or scientific formatting; Subscripts; Superscripts - G06F 18/24133: {Distances to prototypes} - G06F 40/14: Tree-structured documents (parsing G06F40/205; validation G06F40/226) - G06F 40/151: Transformation - G06F 40/211: Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars - G06F 40/216: using statistical methods - G06V 10/82: using neural networks - G06V 30/244: using graphical properties, e.g. alphabet type or font - G06V 30/413: Classification of content, e.g. text, photographs or tables - G06V 30/414: Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text - G06F 16/5846: {using extracted text} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A method, computer system, and a computer program product for automatic equation transformation from text is provided. The present invention may include receiving a text document. The present invention may then include identifying a mathematical formula expressed in the received text document. The present invention may then include removing a plurality of superfluous language from the received text document based on the identified mathematical formula. The present invention may also include transforming the identified mathematical formula into a symbolic representation based on a trained model. The present invention may finally include outputting the symbolic representation. ## Claims ### Claim 1 A method for automatic equation transformation from text, the method comprising: receiving a text document; identifying a mathematical formula expressed in the received text document; removing a plurality of superfluous language from the received text document based on the identified mathematical formula; transforming the identified mathematical formula into a symbolic representation based on a trained model; utilizing a natural language processing technique to generate a tree-like structure using an identified plurality of mathematical equation language, wherein the tree-like structure is generated using an equation dependency parser, an input stack, a processed stack, wherein a neural network with one or more hidden layers is implemented with the equation dependency parser, wherein an input layer of the neural network is represented by a semantic embedding variable, a named-entity recognition variable, a surrounding word variable, and a label embedding vector; and wherein an output layer of the neural network is represented using a softmax function; and outputting the symbolic representation. ### Claim 2 (depends on 1) The method of claim 1, wherein the received text document is a variable length determined by a user. ### Claim 3 (depends on 1) The method of claim 1, wherein identifying the mathematical formula expressed in the received text document further comprises: identifying a plurality of keywords by parsing through a plurality of text associated with the received text document; and comparing the identified plurality of keywords to a database of known mathematical symbols. ### Claim 4 (depends on 3) The method of claim 3, wherein identifying the plurality of keywords by parsing through the plurality of text associated with the received text document comprises utilizing an indexing technique to search the received text document. ### Claim 5 (depends on 1) The method of claim 1, wherein removing the plurality of superfluous language from the received text document further comprises: performing a syntactic analysis of the received text document; and removing a plurality of language from the received text document that is not part of the generated tree-like structure. ### Claim 6 (depends on 1) The method of claim 1, wherein the output mathematical equation is a downloadable text document. ### Claim 7 A computer system for automatic equation transformation from text, comprising: one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage medium, and program instructions stored on at least one of the one or more tangible storage medium for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising: receiving a text document; identifying a mathematical formula expressed in the received text document; removing a plurality of superfluous language from the received text document based on the identified mathematical formula; transforming the identified mathematical formula into a symbolic representation based on a trained model; utilizing a natural language processing technique to generate a tree-like structure using an identified plurality of mathematical equation language, wherein the tree-like structure is generated using an equation dependency parser, an input stack, a processed stack, wherein a neural network with one or more hidden layers is implemented with the equation dependency parser, wherein an input layer of the neural network is represented by a semantic embedding variable, a named-entity recognition variable, a surrounding word variable, and a label embedding vector; and wherein an output layer of the neural network is represented using a softmax function; and outputting the symbolic representation. ### Claim 8 (depends on 7) The computer system of claim 7, wherein the received text document is a variable length determined by a user. ### Claim 9 (depends on 7) The computer system of claim 7, wherein identifying the mathematical formula expressed in the received text document further comprises: identifying a plurality of keywords by parsing through a plurality of text associated with the received text document; and comparing the identified plurality of keywords to a database of known mathematical symbols. ### Claim 10 (depends on 9) The computer system of claim 9, wherein identifying the plurality of keywords by parsing through the plurality of text associated with the received text document comprises utilizing an indexing technique to search the received text document. ### Claim 11 (depends on 7) The computer system of claim 7, wherein removing the plurality of superfluous language from the received text document further comprises: performing a syntactic analysis of the received text document; and removing a plurality of language from the received text document that is not part of the generated tree-like structure. ### Claim 12 (depends on 7) The computer system of claim 7, wherein the output mathematical equation is a downloadable text document. ### Claim 13 A computer program product for automatic equation transformation from text, comprising: one or more computer-readable storage media and program instructions stored on at least one of the one or more tangible storage media, the program instructions executable by a processor to cause the processor to perform a method comprising: receiving a text document; identifying a mathematical formula expressed in the received text document; removing a plurality of superfluous language from the received text document based on the identified mathematical formula; transforming the identified mathematical formula into a symbolic representation based on a trained model; utilizing a natural language processing technique to generate a tree-like structure using an identified plurality of mathematical equation language, wherein the tree-like structure is generated using an equation dependency parser, an input stack, a processed stack, wherein a neural network with one or more hidden layers is implemented with the equation dependency parser, wherein an input layer of the neural network is represented by a semantic embedding variable, a named-entity recognition variable, a surrounding word variable, and a label embedding vector; and wherein an output layer of the neural network is represented using a softmax function; and outputting the symbolic representation. ### Claim 14 (depends on 13) The computer program product of claim 13, wherein the received text document is a variable length determined by a user. ### Claim 15 (depends on 13) The computer program product of claim 13, wherein identifying the mathematical formula expressed in the received text document further comprises: identifying a plurality of keywords by parsing through a plurality of text associated with the received text document; and comparing the identified plurality of keywords to a database of known mathematical symbols. ### Claim 16 (depends on 15) The computer program product of claim 15, wherein identifying the plurality of keywords by parsing through the plurality of text associated with the received text document comprises utilizing an indexing technique to search the received text document. ### Claim 17 (depends on 13) The computer program product of claim 13, wherein removing the plurality of superfluous language from the received text document further comprises: performing a syntactic analysis of the received text document; and removing a plurality of language from the received text document that is not part of the generated tree-like structure.