A.N.I.T.A. (Artificial Neural-Intelligence Translator Assistant) is a vertically integrated artificial intelligence / machine learning solution developed by Stratis, primarily designed to facilitate, partially or completely replace the document-based tasks of a corporate administrative employee.
We regularly expand and develop its functions, while the AI solution and continuously learns throughout the daily collaboration with our customers.
Avoiding expensive and long-term assessment of operational rules, corporate procedures, and programming, A.N.I.T.A. learns them based on patterns so that it can solve complex tasks that could not be solved from a realistic time and financial framework.
A.N.I.T.A.’s machine translation capabilities can be utilized primarily for mass translation needs, or for translation of languages that are rare, making it difficult to provide human translation. Even the most advanced machine-translated solutions cannot be compared to a translation by someone who speaks a particular language on a high level but it is not their aim, neither. It is able to translate massive amounts of documents in such a high quality that their topics and key concepts can be understood, so that they can be searched and the essential parts can be easily selected and then translated with human power in a compelling way.
A.N.I.T.A. is able to continuously execute the translation of thousands of documents without interruption on daily basis, sparing the human workforce and multiplying the amount of material that can be processed in a single day with human power.
Based on words, A.N.I.T.A. can recognize the language of a given sentence or a the entire document with good efficacy, even if the document contains words in other languages (for example, English terms). This is a useful feature for several reasons when processing written documents. On the one hand, typically different workflows apply to documents in different languages (for example, it needs to be taken to a different colleague, or some languages has to be given priority, while others not, etc.). On the other hand, one of the key information in the documents is the language of the document itself. For many processing processes, when translated, or even searched, it is worth knowing the language of the given sentence or document.
Named entity recognition
A.N.I.T.A. can extract important key-expressions from a sentence or document, not (only) based on pre-written dictionaries, but instead on the fundamental artificial understanding of the written text, therefore A.N.I.T.A. can identify a name, location, organization etc. in a text even, if nobody programmed or told her to find that exact person, location, organization etc. before (no need to list all persons, locations, organizations etc.)
Document (text) categorization
A.N.I.T.A. learns on examples, and understands the texts she sees, so it can efficiently determine if a text, document is a particular category. It can also invoke the corresponding processes in the organization and notify the suitable person.
After A.N.I.T.A. mined all the above information of the documents and texts she can scan, all of these are stored in a searchable database. This provides the possibility to search in multiple languages, fuzzy search, search named entities, search for language and category etc.
- Completely flexibly scalable
- A closed system corresponding to the most strictest security principles
- Local installation on demand, not requiring external resources
- Its intelligence can be taught for the needs of our customers, fully customizable
The purpose of anomaly detection is to facilitate the work of professionals detecting unusual or emergency/fraud events from a large amount of data.
Most of the time these processes are either not automated at all or only one source system can be examined at a time. The purpose of machine learning supported anomaly detection is to find correlations between coherent or seemingly independent systems, and odd events that, based on their parameters, do not fit into the operating process, differ from normal operation. Events found by the AI can be examined in detail by a human expert or pre-defined automatisms.
- Ability to collect large and complex datasets (e.g.: log data)
- Creating custom tables (fully customized type, shape, etc. of the input logs)
- Finding anomalies using AI
- Notifying operators
- User friendly and interactive visual interface
- Sufficient teaching data set
- It is suggested to collect data from multiple locations within the organization so that the most complex errors can be detected
- The source systems from which data was collected for teaching must provide data in real time
- It is necessary to fine-tune / re-teach the model periodically for even more accurate results.
- In most cases, the necessary data is available to the company but they are not collected properly.
- This solution can speed up bug fixing processes, furthermore after proper pre-training they can also be alerted before occurring. .