Machine Learning – Soko

Harness the Power of AI for Statutes with Cognitive Data Capture

Mati — Thu, 28 Sep 2023 21:43:04 +0000

data capture

Harness the Power of AI for Statutes with Our Cognitive Data Capture Solution

Soko helps clients stay ahead of financial crimes with their participation reporting service. Their expert team focuses on reducing fraud, criminal activity, and money laundering. With this service, clients can quickly access a person’s involvement in other companies, gaining an advantage in preventing financial crimes.

Technologies:

Algorithms: Transformers, Embeddings, TL-GAN, GAN-based noise, YOLO, Faster R-CNN, NER (Named entity recognition), LSTM

Libraries: Tensorflow, AsanteOCR, Camelot, QR detection, OpenCV, spaCy

Development: Stack MEAN (Mongo, Express, Angular, Node), FastAPI

The Challenge

A team of experts in compliance, technology, risk and data management has the goal of reducing financial crimes and that its clients can protect themselves from fraudsters, criminals, terrorists and money launderers.

For this purpose, the client offers a company participation reporting service. Given the RUT of a natural or legal person, a report is prepared that shows all the companies of which it is (or was) a part and the percentage of participation that corresponds to said person.

The company has a team of 30 people who are in charge of reviewing the history of commercial statutes published in the Official Gazette of Chile and uploading them manually in a web form designed for this purpose. This involves a very repetitive job, subject to errors due to the lexical complexity with which notaries write these documents and entails an inordinate amount of time, since it requires processing the history of 3 million commercial statutes.

The process of drilling in mining consists of obtaining a soil sample by diamond drilling. These samples, which easily reach thousands of feet, are placed in trays intended for this purpose, tabulated and high-resolution photographs are taken. The geologist visually detects and counts fractures, classifying them as natural or induced, depending on whether they are real fractures existing in the earth layers or were caused by drilling or moving the samples.

The Solution

The solution proposed by the Mootech team includes, first of all, the training of multiple machine learning models that involved manually placing labels on each of the existing entities, in a total of 1,000 corporate bylaws.

The project was divided into 3 stages according to the type of document in statutes of creation, statutes of modification and statutes of dissolution so that our data science team could concentrate on specific models.

The development team carried out the implementation of a web application that presents users with the corporate bylaws with the respective entities detected that did not approve the automatic validations. Then the user will be in charge of reviewing them (or updating them if necessary) and thus ensure that erroneous data is not inserted in the database.

Our data scientists selected semantic segmentation algorithms for image processing and model training with the previously labeled data.

CRISP-DM methodology was used and at each completed iteration benchmarking was performed with different experts (geologists) to feed back the model with new training.

Our solution also included the development of a web service to enable interoperability of the proposed system, i.e., to be consumed and integrated by the client entity’s systems. Additionally, we developed scalability elements of the solution through automation and scalability practices known as MLOps, providing capabilities for continuous model retraining and identifying issues that could affect the solution in a production environment.

results

What we achieved

Thanks to the model implemented by Mototech, the client was able to process the history of 3 million corporate statutes in 3 months of execution. 70% of the documents were approved automatically, without requiring manual intervention.

30% of the documents that followed the manual process showed an average statistic of time required for validation or update of 1 minute per document showing the recognized entities.

Geologists have a visual tool that preloads (in less than a second) all the fractures detected in an image and their task is reduced to verify this information in an agile and efficient way.

In addition to the reduction of analysis time, the introduction of the prototype reduced the seniority level required for this task and collaborated in the process of unification of criteria for fracture selection.

In real-world testing, system users report that response capacity and quality of analysis improve significantly, reducing the average time a claim spends waiting to be classified from a week to less than 5 seconds, resulting in greater efficiency in the management process and financial market supervision in general.

A Journey to the Top of the Cloud: SURA’s amazing success in its migration in just 6 months.

Mati — Wed, 30 Aug 2023 00:16:08 +0000

A Journey to the Top of the Cloud:
SURA’s amazing success in its migration in just 6 months

Search, process and analysis of high-volume data in real time with the Soko Solutions Big Data platform.

The Challenge

SURA was operating on local servers (on-premise), which limited its scalability and availability. The company decided to migrate to the native cloud as part of a corporate decision, with the aim of improving efficiency, reducing opera- tional costs and ensuring the adaptability of its services. In addition, they sought to eliminate dependency on WSO2’s costly and highly specialized platform.

The Solution

Our team of experts, Azure Cloud Platform Architects, WSO2 experts, developers and testers, developed a “Specialized Migrator” from WSO2 to Node.js, integrat- ed to Microsoft Azure’s Logics Apps, Azure Functions and Serverless AKS frameworks in record time.

This strategy not only allowed to reduce the entire migration process estimated in 24 months to only 6 months including a rigorous performance testing, but also allows SURA to have a modern and high performance “Flexible Integra- tion” infrastructure with a strong orientation to “Open Insurance”.

Results

Thanks to our solution, SURA achieved significant benefits:

Time and cost savings

The migration to the cloud was completed in only 6 months, which meant savings of more than two years of what a project of this magnitude would take. This change allowed us to improve Time To Market. In addition, by using cloud-native services, there were significant savings in virtual machines and infrastructure costs were reduced by 70%.

High performance and scalability

The cloud-native application developed in Node.js demonstrated extremely high performance, thanks to the best migration practices implemented. This allowed SURA to process large volumes of transactions in an efficient manner, ensuring greater growth capacity for the company.

Expertise

Our team of architects and developers provided SURA with in-depth knowledge of modern architecture and tools, which guaranteed a successful implementation and optimal management of the platform through performance testing in the cloud.

In addition to the migration, we provided additional services to SURA, such as architecture, DevOps and security, to ensure that the company got a complete and customized solution.

Harness the Power of AI for Statutes with Cognitive Data Capture

admin — Sat, 25 Mar 2023 23:40:52 +0000

data capture

Harness the Power of AI for Statutes with Our Cognitive Data Capture Solution

Technologies:

Algorithms: Transformers, Embeddings, TL-GAN, GAN-based noise, YOLO, Faster R-CNN, NER (Named entity recognition), LSTM

Libraries: Tensorflow, AsanteOCR, Camelot, QR detection, OpenCV, spaCy

Development: Stack MEAN (Mongo, Express, Angular, Node), FastAPI

The Challenge

The Solution

Our data scientists selected semantic segmentation algorithms for image processing and model training with the previously labeled data.

CRISP-DM methodology was used and at each completed iteration benchmarking was performed with different experts (geologists) to feed back the model with new training.

results

What we achieved

30% of the documents that followed the manual process showed an average statistic of time required for validation or update of 1 minute per document showing the recognized entities.

Geologists have a visual tool that preloads (in less than a second) all the fractures detected in an image and their task is reduced to verify this information in an agile and efficient way.

Development and Implementation of Claims Management

admin — Sat, 25 Mar 2023 22:05:16 +0000

claims management

Machine Learning Streamlines Claims Management in Financial Market Commission

The Financial Market Commission created an automated system using machine learning to classify claims and inquiries, resulting in significant improvements in response times and efficiency.

The Challenge

The Financial Market Commission (C.M.F), responsible for ensuring the proper functioning, development, and stability of the financial market, is receiving a high and growing number of claims and inquiries from the public, revealing a high inefficiency in response capacity and consequent delay in processing. This entity sought solutions to address this problem by creating an automated system for the structuring and massive processing of claims and inquiries, improving response times significantly, and strengthening its supervisory duties.

The Solutions

To achieve the goal, we developed a system that uses machine learning techniques to classify the claims received by market and entities, products, and matters claimed. This process began with the collection and cleaning of available data, i.e., claims with their corresponding classifications. Then, we used classification algorithms and neural networks to analyze the information, test alternative models and hyperparameter combinations until achieving the expected results.

results

What we achieved

The results of our solution are highly satisfactory. In tests conducted in relevant and simulated environments, we achieved 95% accuracy in classifying claims by market and 89% accuracy in classifying by entities, products, and matters claimed.