Natural Language Processing (NLP) is a core tool for text analytics and mining. It helps enterprises and developers process text efficiently. NLP is widely used in business operations for customers in industries such as e-commerce, entertainment, justice, public security, finance, healthcare, and electric power. NLP will continue to provide more valuable services.
Alibaba Cloud Natural Language Processing includes the following sub-services:
NLP Basic Service provides basic natural language processing services, including word segmentation, part-of-speech tagging, named entity recognition, sentiment analysis, and core word extraction. You can use it for scenarios such as AI chat, chatbots, public opinion analysis, content recommendation, and e-commerce review analysis.
Note: NLP Basic Service 1.0 has been unpublished. You can use NLP Basic Service 2.0 instead.
NLP Self-Learning Platform is an adaptive annotation, training, and service platform for industry-specific NLP. It is designed for users with little or no background in algorithms. The platform supports custom algorithms for text entity extraction, text classification, key phrase extraction, sentiment analysis, relationship extraction, short text matching, and product review parsing.
Enterprise Intelligent Search provides offline data processing and search service API operations as a Platform as a Service (PaaS). It supports deployment on the public cloud, Apsara Stack, hybrid cloud on cloud-native infrastructure, and in private environments. The service evolves from relevance-based search to cognitive intelligence by integrating semantics and knowledge into the search process and results. This provides an efficient and highly accurate search product.
NLP Basic Service
Multi-language word segmentation
This service provides intelligent word segmentation. It is developed by a professional team that continuously updates the data and models. You can simply call the relevant API operations to retrieve the results. The service currently supports Simplified Chinese, English, and Thai.
Part-of-speech tagging
You can use the part-of-speech tagging service to quickly assign a part of speech to each word. When combined with the word segmentation service, you can perform deeper text mining without worrying about new word discovery or disambiguation.
Named entity
The named entity recognition service helps you quickly identify entities in text and discover the relationships between them. It is an essential tool for deep text mining, knowledge base construction, and other common natural language processing tasks.
Sentiment analysis
This service leverages massive amounts of big data to perform sentiment analysis. It analyzes short texts to determine whether the sentiment is positive, negative, or neutral. It has broad applications in business areas such as public opinion monitoring, topic moderation, and reputation analysis clustering.
Core word extraction
This feature is trained on massive amounts of data, including core words from e-commerce titles and categories. It calculates a relevance score for each word to measure its connection to the sentence, and then identifies and extracts the core words. It is suitable for extracting core words from e-commerce search queries, titles, and other short texts, typically those with fewer than 25 words.
Intelligent text classification
This feature automatically maps a piece of text to a specific category to help you quickly classify text. It also identifies and extracts key labels from the text. The feature supports both flat and hierarchical, tree-structured category systems. Two default classification systems are built-in and ready to use: one for the news and information realm and one for the e-commerce realm.
Text information extraction
This feature is primarily designed for contract extraction. It uses AI technology to automatically read documents and extract key information. This simplifies mechanical and repetitive tasks and helps businesses with document review and data entry.
Product review parsing
Product review parsing analyzes reviews and comments from consumers. It can also analyze short, conversational text, such as social media posts. This feature is not suitable for long news articles.
NLP Self-Learning Platform
The NLP Self-Learning Platform is an adaptive annotation, training, and service platform for industry-specific NLP. It is designed for users with little or no background in algorithms. The product supports custom algorithms for text entity extraction, text classification, key phrase extraction, sentiment analysis, relationship extraction, short text matching, and product review parsing. You do not need an extensive background in algorithms. You can simply annotate or upload a suitable amount of data to the platform to obtain a high-quality NLP model.
Enterprise Intelligent Search
Enterprise Intelligent Search from Alibaba DAMO Academy is built on DAMO Academy's extensive experience in natural language processing. It focuses on unified enterprise search scenarios and provides precise search across multiple, diverse sources. The service offers offline data processing and search API operations as a PaaS. It supports deployment on the public cloud, Apsara Stack, hybrid cloud on cloud-native infrastructure, and in private environments. The product uses natural language processing, machine learning, and internal enterprise knowledge bases. It evolves from relevance-based search to cognitive intelligence by integrating semantics and knowledge into the search process and results. This provides an efficient and highly accurate search product that helps users find complete and precise results. For enterprise customers, it offers features such as interactive multi-turn conversational search, address book search, address search, and document search. It also provides retrieval-augmented generation for scenarios that involve interactions between enterprises and Large Language Models (LLMs).