Our company creates solutions that will enhance your business with cognitive technologies of IBM Watson.
IBM Watson is a cognitive system that allows human and computers partner in a new way.
It is a technology using Natural Language Processing and Machine Learning to enable natural interaction between human and computing systems and improve human knowledge abilities in various fields.
Key services and products:
Conversation is a service for building and deployment of chatbots that understand natural language and respond in it, in various channels, including mobile devices, messaging platforms, and even robots.
Watson Virtual Agent allows deployment of a pre-trained digital agent with industry and domain knowledge quickly and without significant investment in training or skills. Also, it has an Engagement Metrics feature to view analytics about the interactions with Watson Virtual Agent and to improve customer service even more.
Watson Discovery is for analysis of structured and unstructured data, be it your own proprietary data, public or third-party data. A set of integrated, automated Watson APIs enables it to crawl, convert, enrich and normalize data, and perform natural language processing tasks.
Natural Language Understanding service allows analyzing unstructured text and extracting meta-data such as keywords, concepts, categories, entities, relations, sentiment, emotion. It can be easily customized for a certain domain or industry with custom models using another Watson service, Knowledge Studio. It is used for social media monitoring, or advertising optimization, content recommendation, etc.
IBM Watson Knowledge Studio serves to create
- a machine-learning model with understanding of a specific industry or
- a rule-based model to find entities in documents.
It provides tools for annotating unstructured domain literature, and, basing on the annotations, creates a custom machine-learning model that understands the linguistic nuances, meaning, and relationships specific to the domain. After testing, the finished machine-learning model can be deployed to other Watson cloud-based services and products.
Discovery News is a dataset of English-language continuously updated news sources that can be used for sentiment analysis, to detect events and trending topics, create news alerts with the support for keywords, entities, categories, etc.
Document Conversion allows converting documents to different file types, works in multiple languages.
IBM Watson Speech to Text service offers real-time speech recognition transcribing audio from 7 languages with the options of recognizing different speakers in audio and spotting particular keywords.
IBM Watson Text to Speech service allows converting text into natural-sounding audio in various voices and streaming the results back with minimal delay. Supports various languages and dialects.
Language Translator enables texts translations with high-level personal data protection. Domain-specific models and customization options help to make the translations more precise. It is widely used to enhance communications with customers in their language.
Natural Language Classifier is a service for analyzing short text inputs, such as a sentence or phrase, returning best matching predefined classes for them. A classifier has to be trained on example data. It can be used, for instance, in Support service, to understand and sort requests by urgency and also, to define, to whom or to which departments to send them.
Personality Insights is used to analyze email, blogs, tweets, and forum posts and predict personality characteristics, values and needs, which is important to personalize the customer’s experience.
Tone Analyzer allows understanding emotions from the text which helps to improve communication with customers.
Examples and Use Cases
Virtual Agent
A company that wants a human-like virtual agent to provide Support Service and to help with the self-service to its customers deploys Watson Virtual Agent that already has knowledge in its industry. The company needn’t spend much time training the agent. Watson Virtual Agent already understands common queries because it has been pre-trained on frequently occurring client intents, the real purpose of the many different ways that a customer can ask a question. So we need just to add specific company info and launch the Agent.
IBM Watson NLU+Knowledge studio+Document Conversion
Recruiters constantly analyze immense quantity of CV and resumes. We use Knowledge studio to train the system, Document Conversion to convert CV from doc(x)/pdf/odt and IBM Watson NLU to analyze them. In this way, we extract required entities from resume and leave for requiter’s review only those which meet some initial requirements.
Speech to text+Natural Language Classifier
Users tell the chatbot about the price, the flavor type, and other preferences in wine, which is converted to text by means of Speech to text API, analyzed and then Natural Language Classifier takes care of providing people with a list of wines that will suit them considering their wishes.