Arabic, International & Multilingual Software Desktop Publishing Machine Translation Document Management OCR ASR TTS MultimediaA

Multilingual Information Retrieval

Concept
With the Internet skyrocketing to become the fastest growing media for information exchange, the number of non-English sites and users are also rapidly growing. Consequently, the wealth of information in various languages is also increasing. The need to access this wide information base among English speaking users is also rising. On the other hand the abundance of English information on the Internet created a need for non-English speakers, to use this information in their native languages.

Morphology Analyzer.
A particular feature of this system is its multilingual Morphology Analyzer. The morphology analyzer is capable of analyzing and generating all inflectional paradigms for all classes of verbs, nouns and adjectives or any other part of speech. The module is also capable of reducing character strings to their dictionary look-up form (root or stem, as required). The analyzer processes stems as well as meaningful root derivational morphology.

Query Translation
TextFinder™ has bilingual dictionaries that allow users to search in multilingual text simultaneously. The user can type a search word or phrase in one language and retrieve the matching words with successive part of speech in other languages using the morphology analyzer as well as the large-size dynamic bilingual lexicons.

Proper Noun Extraction (Name/Event Search)
AppTek has developed a Proper Noun Extraction System that is used within TextFinder™. The system can identify all proper nouns in multiple languages such as human names and geographic countries, cities, places, etc., utilizing the large names database developed by AppTek as well as surface parsing and statistics. The search can also be done using exact match and fuzzy match techniques. It allows for Name/Event searches.

Thematic Search
AppTek has developed bilingual Thematic and Thesaurus Lexicons that will facilitate the search for a specific topic within text information or similar meanings for a word. User can search for a topic theme like “crime” and be able to retrieve text that includes words like robbery / homicide / kidnapping /, etc. This search can be associated with human, geographic, proper nouns of choice etc. Statistical meaningful associations are also available.

User Dictionaries
TextFinder™ has a Dictionary Maintenance Utility (DMU) allowing the user to create user-specific glossary, keywords list and special terminology that can be used for future search.

Indexing Manager
The indexing component utilizes TextFinder™ linguistic components to build a stem-based index for the source text that will allow users to search using TextFinder™ advanced features. The source text can be files with various formats, database such as Oracle or SQL or third party applications using TextFinder™ API.

Search Manager
The search manager is used to manage search queries and enable users to utilize the linguistic features. The search manager has a full-featured HTML parser that supports built-in variable management for parsing and presenting HTML forms; the search manager enables users to customize their own web interface for Text Finder™ or integrate TextFinder™ features with third party applications.

Application Programming Interface
TextFinder™ can work as a turnkey search engine or can integrate with third party search engines or applications through its API to enhance their search capabilities using TextFinder™ linguistic features.

Functions
TextFinder™ is a search engine as well as electronic publishing software, which hyper-links to graphs, figures and other information. The target documentation as well as the source can be viewed electronically and in synchronization. The software is equipped with linguistic intelligence to provide powerful search features. Users can search for desired information easily and more accurately. Users can also create advanced search queries for very specific information based on combinations of search parameters. With a query translation capability users can also retrieve data in multiple languages. Users can type a query in one language and search against another language. Search results can be translated using AppTek Machine Translation System (TranSphere), which can be integrated with TextFinder™. TextFinder™ is available on various architectures, including Web-based, client/server or stand-alone. TextFinder™ runs under Windows operating system as well as Unix. TextFinder™ interfaces with ICR & OCR software to digitize source material. The system can be integrated with AppTek machine translation, database systems, document management, content and knowledge management, electronic multimedia publishing tools and other Windows applications.

End-user Products
Indexing Server running on B-tree Database, Oracle or SQL
Search Server
Search and Indexing Query Expander
Front-end Search Manager
Third Party Integration Tools

Please, contact George N. Hallak for Multi-License pricing and customized solutions