PERCENT Intelligent Text Analysis System focuses on the text-based natural language processing technology and meets the business needs of government and enterprise clients of natural language processing with one-stop services by making use of rich data mining, machine learning, and artificial intelligence algorithms to train online and offline semantic models. It aims to improve its scientific decision-making ability and efficiency, and improve social and economic benefits.
The system supports localized and distributed deployments and can rapidly expand the quantity of nodes based on actual data scale. The nodes can be back-up copies for each other, which guarantees sound operation of the system.
The system supports basic text analysis in 10 languages, including Chinese, English, German, French, Russian, Greek, Danish, Finnish, Spanish, and Portuguese.
Practical experience of over 1,000 clients involving industries such as finance, 3C manufacturing, automobile, healthcare, TV and movies, entertainment, and Apps has helped accumulate a rich industry lexicon.
The textual model trained using deep learning can significantly improve the effectiveness of textual analysis and support training with small-scale tagged data, rapidly meeting the business demands of different industries.
The system supports localized and distributed deployments and can rapidly expand the quantity of nodes based on actual data scale. The nodes can be back-up copies for each other, which guarantees sound operation of the system.
The system supports basic text analysis in 10 languages, including Chinese, English, German, French, Russian, Greek, Danish, Finnish, Spanish, and Portuguese.
Practical experience of over 1,000 clients involving industries such as finance, 3C manufacturing, automobile, healthcare, TV and movies, entertainment, and Apps has helped accumulate a rich industry lexicon.
The textual model trained using deep learning can significantly improve the effectiveness of textual analysis and support training with small-scale tagged data, rapidly meeting the business demands of different industries.
The system supports localized and distributed deployments and can rapidly expand the quantity of nodes based on actual data scale. The nodes can be back-up copies for each other, which guarantees sound operation of the system.
The system supports basic text analysis in 10 languages, including Chinese, English, German, French, Russian, Greek, Danish, Finnish, Spanish, and Portuguese.
Practical experience of over 1,000 clients involving industries such as finance, 3C manufacturing, automobile, healthcare, TV and movies, entertainment, and Apps has helped accumulate a rich industry lexicon.
The textual model trained using deep learning can significantly improve the effectiveness of textual analysis and support training with small-scale tagged data, rapidly meeting the business demands of different industries.
On the basis of industry applications and business scenarios, the system can judge the sentiment tendency of textual data as positive, negative, or neutral, and identify and extract the word-of-mouth information with clear assessment orientation, such as positive or negative evaluations of each dimension of the products.
Categorization refers to the ability of the system to automatically categorize a text as per the semantic information contained in the text based on the existing category system. Clustering is the system's ability to automatically cluster a mass of textual data into multiple classes according to the internal data distribution and semantic features of the texts, and give descriptive keywords to each class of data.
The system automatically extracts some representative keywords from a text as the descriptive tags of the text. Based on this, it can further extract some representative expressions as the summary, forming a succinct and summarized description of the textual data.
It intelligently identifies the named entity information contained in the text, including personal names, location names, time, names of organizations, and technical terms, etc.
It supports precise word segmentation of Chinese text contents and intelligently judges the class of the corresponding word, providing basic support for subsequent analysis.
On the basis of industry applications and business scenarios, the system can judge the sentiment tendency of textual data as positive, negative, or neutral, and identify and extract the word-of-mouth information with clear assessment orientation, such as positive or negative evaluations of each dimension of the products.
Categorization refers to the ability of the system to automatically categorize a text as per the semantic information contained in the text based on the existing category system. Clustering is the system's ability to automatically cluster a mass of textual data into multiple classes according to the internal data distribution and semantic features of the texts, and give descriptive keywords to each class of data.
The system automatically extracts some representative keywords from a text as the descriptive tags of the text. Based on this, it can further extract some representative expressions as the summary, forming a succinct and summarized description of the textual data.
It intelligently identifies the named entity information contained in the text, including personal names, location names, time, names of organizations, and technical terms, etc.
It supports precise word segmentation of Chinese text contents and intelligently judges the class of the corresponding word, providing basic support for subsequent analysis.
On the basis of industry applications and business scenarios, the system can judge the sentiment tendency of textual data as positive, negative, or neutral, and identify and extract the word-of-mouth information with clear assessment orientation, such as positive or negative evaluations of each dimension of the products.
Categorization refers to the ability of the system to automatically categorize a text as per the semantic information contained in the text based on the existing category system. Clustering is the system's ability to automatically cluster a mass of textual data into multiple classes according to the internal data distribution and semantic features of the texts, and give descriptive keywords to each class of data.
The system automatically extracts some representative keywords from a text as the descriptive tags of the text. Based on this, it can further extract some representative expressions as the summary, forming a succinct and summarized description of the textual data.
It intelligently identifies the named entity information contained in the text, including personal names, location names, time, names of organizations, and technical terms, etc.
It supports precise word segmentation of Chinese text contents and intelligently judges the class of the corresponding word, providing basic support for subsequent analysis.
The system can monitor and analyze users’ evaluations and complaints about the whole business chain including enterprise brands, products, and services by integrating enterprise customer service, collecting Internet data, and other multi-source heterogeneous data, further providing scientific and efficient decision-making basis for enterprise business activities to be optimized and improved.
The system supports internet panoramic monitoring of public opinions and business information, including industry information, negative public opinions, trending events, brand monitoring, competing product analysis, and marketing tracking, etc., and helps clients to comprehensively understand their and competitors' public opinions and business information, thus allowing clients to take measures proactively and respond rapidly and accurately.
The system can provide intelligent subject selection, aided verification, convenient article release, and other business support by collecting clients' internal and external data resources with a focus on collecting, editing, and release of media articles, thus enabling media clients to improve their core business competitiveness.
Based on enterprises' internal data and internet-related data, the system can mine enterprise knowledge and information, build complete enterprise knowledge systems, and provide professional knowledge support for product research & development, marketing, and intelligent question & answering.
The system can monitor and analyze users’ evaluations and complaints about the whole business chain including enterprise brands, products, and services by integrating enterprise customer service, collecting Internet data, and other multi-source heterogeneous data, further providing scientific and efficient decision-making basis for enterprise business activities to be optimized and improved.
The system supports internet panoramic monitoring of public opinions and business information, including industry information, negative public opinions, trending events, brand monitoring, competing product analysis, and marketing tracking, etc., and helps clients to comprehensively understand their and competitors' public opinions and business information, thus allowing clients to take measures proactively and respond rapidly and accurately.
The system can provide intelligent subject selection, aided verification, convenient article release, and other business support by collecting clients' internal and external data resources with a focus on collecting, editing, and release of media articles, thus enabling media clients to improve their core business competitiveness.
Based on enterprises' internal data and internet-related data, the system can mine enterprise knowledge and information, build complete enterprise knowledge systems, and provide professional knowledge support for product research & development, marketing, and intelligent question & answering.
Inquiry