WaveAccess ValueAI

The main articles are:

• CRM (Customer Relationship Management System)
• Machine Learning

2024

Add RAG Search

The developer of the ValueAI platform on September 30, 2024 presented an additional business opportunity - an intelligent search for corporate knowledge bases based on RAG (Retrieval Augmented Generation). This technology combines the generative capabilities of AI with real-time search of relevant sources of information, allowing reliable and accurate responses. This approach reduces data analysis time, minimizes errors, optimizes costs, and improves communication efficiency. It is also possible to use on-premium AI models.

The ValueAI platform acts as a ready-made tool for RAG search, which combines a knowledge base loaded through an interface or integrated by, API conversion files to a search-appropriate format, search for the relevant, information its processing by an AI model and response generation. chat boat

To interpret the information found, the RAG module in ValueAI uses one of the large language models YandexGPT, ChatGPT or LlaMA. It is also possible to use on-premium models.

The module can also be configured to independently access external updated enterprise systems to automatically replenish the database with new records, as well as adapt it to work with specialized subject areas that require deep contextual understanding.

Searching with RAG significantly improves the work of support services, contact centers and HR departments. We are seeing a special interest in this function from a business working with large amounts of information, in particular government agencies, production, law firms, pharmaceuticals, fintech and construction. Also, RAG search is in demand among organizations with high staff turnover or, conversely, intensively increasing staff, and large companies with a deployed customer support service, - said Alexander Azarov, General Director of VeivAxess, a ValueAI platform developer.

YandexGPT 3 Availability

Support feature YandexGPT 3 is available on the ValueAI platform. Now companies can fully use the potential of the generative text model Yandex to solve complex business and work tasks with. data This was WaveAccess (WaveAxess) announced on September 18, 2024.

Now the business can strengthen its solutions and the development of new AI products with the power of LLM by accessing it via the API from the ValueAI interface.

The ValueAI platform can be deployed both in the cloud and on-premium - with the ability to access YandexGPT and other third-party services.

The updated ValueAI module not only improves training of AI models through data enrichment, but also offers the ability to separately use YandexGPT for chats and document processing. The synergy of ValueAI and YandexGPT solves a number of problems in which there is initially insufficient data, offering a ready-made solution out of the box. This opens up the ability for businesses that previously had limitations in launching machine learning projects to integrate AI. Also, this approach is relevant for companies with high information privacy requirements and for those who want to use LLM to quickly test business hypotheses.

Using ValueAI to work with YandexGPT in AI projects allows you to:

• Avoid privacy problems thanks to the ability to depersonalize data;
• Provide control over the interpretability and transparency of the results, minimizing errors in the "output" data;
• Solve the problem of preparing and rechecking the completeness and quality of the dataset for AI training;
• It is convenient to integrate the AI model into existing business processes and systems.

Machine learning models, as part of AI, analyze large amounts of information, identify patterns and make predictions. LLMs, on the other hand, transform the obtained complex conclusions into an understandable and convenient form for making decisions. The combination of ValueAI and YandexGPT opens up many new applications of AI in business, allowing not only to analyze data in depth, but also to quickly apply the knowledge gained in practice. At the same time, ValueAI allows you to bypass the pitfalls associated with the use of LLM in AI projects, simplify integration and take care of personal data, "said Alexander Azarov, CEO of VeivAxess.

2023: Adding the OCR feature

On April 3, 2023, WeivAxess announced that it had updated the functionality of ValueAl, a tool for processing incoming requests based on artificial intelligence. The OCR (Optical Character Recognition) feature has been added to the platform. Now the AI solution can recognize scans of documents and convert them to Docx format.

ValueAl is a Russian-designed tool that allows you to automate the processing of any incoming online applications (emails, messages to instant messengers, tickets to the support service, etc.). The product solves the tasks of classifying, prioritizing and routing requests. WaveAxess improved it by adding a function that can convert scanned text documents, photos of documents and tables into a machine-readable text format.

The implemented Optical Character Recognition (OCR) technology on the ValueAI platform solves problems associated with manual document processing and data entry, reducing errors and optimizing labor costs by freeing specialists from routine work.

The system detects images inside the text file, and also gives the user appropriate prompts. Both individual files and document archives can be downloaded. It is also possible to download and upload files via the API, which allows you to automate work with data. The solution can be integrated with e-mail, chat bots, CRM and any other sources of information.

We constantly analyze the needs of the market and our customers and introduce demanded functions and technologies into our solutions. Any business implies the presence of documentation on paper, and this format requires a lot of time for processing. The emerging ValueAI function is designed to automate the process of transferring data to electronic format. This will positively affect the speed of work processes, - said Alexander Azarov, General Director of VeivAxess.

For April 2023, the OCR converter can work with documents in Russian and English, WeivAxess plans to add the ability to recognize texts in German. The main emphasis in the development of the function will be aimed at processing the information received into a data set for further classification. It is expected that in the future - with the introduction of the Named Entity Recognition (NER) function

ValueAI will be able to find specified categories of words and phrases in the text, which will improve the process of classifying applications.

2021

Add the ability to integrate data from chatbots and CRM systems

In November 2021, an update to ValueAI, a solution for processing incoming requests using artificial intelligence, was presented. The developer, IT company WaveAccess, has added the ability to integrate data from chat bots and CRM systems. Prioritized requests and ValueAI recommendations for their processing are available to the user in a single CRM interface.

For this, SyncIT is used - a platform for migration data and (systems integration WaveAccess product). Connection to chat bots and CRM is carried out through REST. API

Text requests from clients sent to e-mail from chatbots and directly are uploaded to CRM using the dedicated SyncIT integration application and then transferred to ValueAI for processing. SyncIT then returns the data to CRM, where the target field automatically reflects the received ValueAI forecast. For example, based on the conclusions made, the model can recommend which cases should be considered first, sorting them according to the criterion of urgency or margin.

In ValueAI, the model is initially trained on the datacet, which is also unloaded from CRM using SyncIT: applications from email and chat bots previously manually processed by specialists, and the results of their processing.

Now the user in the CRM interface sees not only systematic customer requests, but also recommendations for their high-quality "closure." A simplified system for interacting with tools through a single work window allows you to reduce errors, reduce the proportion of lost requests, reduce response time, increase customer loyalty, and optimize contact center costs.

Many companies use chatbots as the first line of customer support - integration with this tool in conjunction with CRM has become a logical stage in the development of ValueAI. SyncIT allowed us to gracefully combine all elements of the integrated ecosystem. Now we can offer users not only the benefits of artificial intelligence, but also a kind of optimized sales tool, "said Alexander Azarov, CEO of WaveAccess.

ValueAI is the best way to introduce artificial intelligence into business processes. The ready-made intelligent tool allows you to automate the processing of incoming calls: emails, messages in instant messengers, requests for the supply or provision of services, tickets to technical support or requests for detailed information. This is the only completely Russian solution with an interface in Russian and Russian-language support.

In terms of updates scheduled for the end of Q4 2021: advanced visualization of data analytics, support for processing different languages, processing photos and scans of documents, extracting text from images, automatic selection of the best model.

Inclusion of ValueAI in the register of Russian software

On October 15, 2021, WaveAccess announced that its ValueAI AI solution, an intelligent tool for automating incoming messages, was included in the Unified Register of Russian Programs for Electronic Computers and Databases. 

The solution fully meets the requirements of the legislation of the Russian Federation, which prescribes the federal authorities, state corporations, state and municipal customers to purchase software included in the register as part of the import substitution program. 

ValueAI teaches its own AI

ValueAI is a machine learning tool that allows you to automate the processing of incoming requests from users or customers: emails, messages to instant messengers, requests for the supply or provision of services, tickets to technical support or requests for detailed information. The system actually replaces the first line of the client service, and conclusions from the data with its help can be extracted without writing a line of code.

ValueAI was created to solve a number of problems, among which the classification of incoming requests (and their automatic transfer for processing to the corresponding department), reducing the waiting time for a response to requests, as well as reducing the proportion of unanswered requests and lost requests are distinguished.

AI is trained on the customer's own incoming requests (applications with text and/or numerical data). Knowledge in the field of machine learning or programming is not required to use the system - all operations are performed in a visual editor. ValueAI is implemented in any organization within two weeks.

"The inclusion of ValueAI in the registry is a significant event for us. Now our product is recommended for implementation in public sector organizations, where automation of processing of online applications requiring sorting and analysis, with a large flow of requests, can significantly increase the speed and quality of services to the population. Including due to the prompt identification of applications that must be submitted for processing as a priority, "- noted by Alexander Azarov, WaveAccess CEO.

For sensitive information customers, ValueAI enables you to integrate your solution into your organization's on-premium framework, avoiding the risks of migrating data to the cloud. Installation, configuration and startup of the system are carried out with the support of the developer. At the same time, customization for specific customer needs is possible. 

ValueAI View

WaveAccess, a Russian custom software developer, introduced ValueAI on April 28, 2021, a machine learning tool that allows you to automate the processing of incoming requests from customers or users: emails, messages to instant messengers, applications for the supply or provision of services, tickets to technical support or requests for detailed information. The system actually replaces the first line of the client service, and conclusions from the data with its help can be extracted without writing a line of code.

ValueAI was created to solve a number of problems, among which the classification of incoming requests (and their automatic transfer for processing to the corresponding department), reducing the waiting time for a response to requests, as well as reducing the proportion of unanswered requests and lost requests are distinguished.

Thus, the system can be used by both sales and marketing departments (to find relevant requests and increase conversion), as well as technical support services and quality departments (to improve the quality of service, reduce costs and improve SLA).

If your company has a large stream of online applications that require manual sorting and analysis, automation of their processing can become a point of business growth. Automatic processing of incoming requests is a way in a large stream of requests to highlight the most "high-quality" and important ones according to the specified criteria, and transfer them to processing in priority order. ValueAI allows you not only to configure automatic lead processing, but also to extract from applications valuable conclusions and insights about processes, clients, the quality of your services, etc., which are not available earlier, "said WaveAccess CEO Alexander Azarov.

With ValueAI, any company can safely train and implement AI in its organizational software within two weeks. Knowledge in the field of machine learning or programming is not required to use the system - all operations are performed in a visual editor.

AI is trained on the company's own incoming requests (applications with text and/or numerical data). ValueAI helps you prepare a data set (data set), clear it, train a model on it. Once accurate, the solution is ready to integrate with the organization's REST API software. As a result of the implementation, the business will be able to prioritize applications by urgency or based on their margin, as well as sort them by department.

Integration can be done in the on-premium framework of an organization to avoid the risks of migrating data to the cloud. Another good quality of ValueAI is that it is a completely Russian product (as of April 2021, it is being registered in the register of domestic software), so it can be used by any customers as part of the import substitution program .

Installation, configuration and startup of the system are carried out with the support of the developer. At the same time, customization for specific customer needs is possible.

ValueAI is provided under the license model: a one-time license payment in the first year and 20% of the license price from the second year for using the product.