Oracle International Corporation

Jun Qian

Tao Sheng

Iman Zadeh

Amit Agarwal

Kulbhushan Pachauri

Mengqing Guo

Yazhe Hu

Jinghou Zhang

Mohammadhossein Chaghazardi

Liyu Gong

Fuheng Wu

Zhonghai Deng

Jeaff Wang

Arunjeyan T V Seshier Venkatachalapathy

Wenjing Yang

Yuying Wang

Ivan Dimitrov Davchev

Sofiane Ouaguenouni

Olaitan Olaleye

Techniques are described for HTML-based image generation. An example, method can include generating hypertext markup language (HTML) code for a table comprising a table structure of a set of rows and columns. The method can further include generating HTML code for a text to populate a cell of the table. The method can further include generating a rendered image of the table using the HTML code. The method can further include detecting a first pixel of the rendered image comprising the first color, and a second pixel of the rendered image comprising the second color. The method can further include detecting the text on the rendered image. The method can further include generating a bounding box, surrounding the detected text. The method can further include generating annotation comprising a bounding box parameter and a text parameter.

Synthetic table generation pipeline for training deep table extraction models

Automated techniques are for generating a large volume of diverse training data that can be used for training machine learning models to extract KV pairs from document images. Given a single input document image and associated annotation data, a large number of diverse synthetic training datapoints are automatically generated by a synthetic data generation system, each datapoint including a synthetic document image and associated annotation data. The generated synthetic training datapoints can be used to train and improve the performance of ML models for extracting KV pairs from document images. In certain implementations, multiple synthetic datapoints are generated by varying the values associated with a key for a content item within the input document image.

Generating synthetic training data including document images with key-value pairs

Techniques for multi-stage training of a machine learning model to extract key-value pairs from documents are disclosed. A system trains a machine learning model using a set of training data including unlabeled documents of various document categories. The initial stage identifies relationships among tokens, or words, numbers, and punctuation, in documents. The system re-trains the machine learning model using a set of training data which includes a particular category of documents while excluding other categories of documents. The second training stage is a supervised machine learning stage in which the training data is labeled to identify key-value pairs in the documents. In the initial training stage, the system sets parameters of the machine learning model to an initial state. In the second stage, the system modifies the parameters of the machine learning model based on the characteristics of the training data set including the documents of the particular category.

Multi-stage machine learning model training for key-value extraction

The present embodiments relate to a language identification system for predicting a language and text content of text lines in an image-based document. The language identification system uses a trainable neural network model that integrates multiple neural network models in a single unified end-to-end trainable architecture. A CNN and an RNN of the model can process text lines and derive visual and contextual features of the text lines. The derived features can be used to predict a language and text content for the text line. The CNN and the RNN can be jointly trained by determining losses based on the predicted language and content and corresponding language labels and text labels for each text line.

Vision-based document language identification by joint supervision

Techniques are disclosed for dynamic time-based custom model generation as part of infrastructure-as-a-service (IaaS) environment. A custom model generation service may receive a set of training data and a time-based constraints for training a machine learning model. The custom model generation service may subsample the training data and generate a set of optimized tuned hyperparameters for a machine learning model to be trained using the subsampled training data. An experimental interval time of training is determined and the machine learning model is trained on the subsampled training data according to the optimized tuned hyperparameters over a set of training intervals similar to the experimental time interval. A customized machine learning model trained in the time-based constraint is output. The hyperparameter tuning may be performed using a modified mutating genetic algorithm for a set of hyperparameters to determine the optimized tuned hyperparameters prior to the training.

Techniques for dynamic time-based custom model generation

A computing device may access a target code for implementing an application. The device may identify addresses for one or more functions or one or more variables associated with the target code. The device may generate an interval tree comprising a root node and one or more function nodes. The device may in response to the target code invoking a function or variable: generate an intercept function configured to intercept communication between the target code and a call address for the at least one of the one or more functions or the one or more variables invoked by the target code. The device may intercept data communicated between the target code and the call address. The device may store the intercepted data as a function node in the interval tree. The device may transmit the interval tree to a user device.

Application performance monitoring for monolithic applications and distributed systems

A computing device may receive a set of user documents. Data may be extracted from the documents to generate a first graph data structure with one or more initial graphs containing key-value pairs. A model may be trained on the first graph data structure to classify the pairs. Until a set of evaluation metrics for the model exceeds a set of deployment thresholds: generating, a set of evaluation metrics may be generated for the model. The set of evaluation metrics may be compared to the set of deployment thresholds. In response to a determination that the set of evaluation metrics are below the set of deployment thresholds: one or more new graphs may be generated from the one or more initial graphs in the first graph data structure to produce a second graph data structure. The first and second graph can be used to train the model.

Techniques for graph data structure augmentation

Described herein are systems and methods for identifying the similarity between language characters. As described herein, a pair of language characters is received at a language character match engine. The language character match engine is adapted to receive encoding configuration information from each of a plurality of encoding components, and is adapted to encode the pair of language characters based on the unique structure of each language character to generate a pair of string identification characters for each encoding component. Thereafter, each pair of string identification characters is compared to one another to generate a similarity score, and the similarity score for each pair of string identification characters is combined to create a composite similarity score. The composite similarity score represents a similarity between the pair of language characters, and is used to identify the similarity between the pair of language characters.

Growing community of inventors

Pasadena, CA, United States of America

Jun Qian