Características de nuestro programa
Descripción del programa
Se trata de un programa de reconocimiento de significado. Funciona sobre cualquier contenido digital (en los formatos más comunes del mercado txt, doc, epub, pdf) con independencia de su idioma (funcionaría inicialmente en inglés, español y chino, pero está previsto hacerlo en otras lenguas).
Qué detectamos en el contenido
Características de nuestro programa. El programa reconoce:
- El significado de palabras individuales en cualquiera de sus formas gramaticales (de acuerdo al contexto en el que aparecen)
- El vocabulario base de un documento digital y busca las definiciones clave de dicho vocabulario
- Traducciones de dichas palabras en las lenguas preseleccionadas
- Localidades
- Composiciones musicales
- Obras literarias
- Obras artísticas de diversa índole
- Películas
- Productos comerciales de todo tipo
- Traduce obras o documentos y revela la fuente de origen
- Entidades no especificadas (fórmulas matemáticas, sistemas métricos, de peso, etc.) y permite hacer conversiones entre estas
- Relaciones entre documentos en distintos formatos (sean estos textos, archivos musicales, imágenes o videos)
- El “sentiment analysis” de las entidades en redes sociales
- Reconoce contenidos en las Api de terceros (Dropbox, Google Cloud, Google Movie, Amazon, Itune, Spotify, etc.)
- El área temática de los contenidos
- En la redes sociales a personas que hablan de contenidos semejantes entre sí
- Búsquedas semánticas dentro de los contenidos que ha procesado
- Hace un clustering del documento (a que área pertenece [ej., “economy, business, finance, labour” 70%, “education” 30%])
- Y a partir de ahí establece una especie de “Page Rank” de las fuentes de información que alimentan las ventanas, que permite seleccionarlas y prioriza unas sobre otras a partir de los feedback de los usuarios. Por ejemplo, si eres médico prefieres que las referencias médicas venga de tu cuenta en http://www.uptodate.com y si eres abogado de una base de datos especializada. Si eres músico prefieres que sea el Billboard, etc., etc.
- Nuestro programa reconoce productos comerciales y genera “Playlist” o “Wishlist” de estos utilizando las API de las principales tiendas Online, de manera que puedan ser adquiridos con un click. Por ejemplo, analiza una receta de cocina y genera un shoping cart con todos los ingredientes, permitiendo al usuario comprarlos en Wallmart.
Características de nuestro programa
Patente provisional
EFS ID: 18628712
Application Number: 61972699
Confirmation Number: 2126
Title of Invention: ONLINE ANNOTATION GENERATION AND DISPLAY SYSTEM AND METHOD
CROSS-REFERENCE TO RELATED APPLICATIONS
- Not Applicable.
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
- Not Applicable.
INCORPORATION BY REFERENCE OF MATERIAL SUBMITTED ON A COMPACT DISC
- Not Applicable.
Technical Field
- The technical field relates generally to information systems and, more specifically, to the dissemination of data garnered from network-accessible databases and servers.
Background
I
With the advent of the Internet and increasingly powerful computing devices, users and entities have uniformly moved to consume all or most of the information they require online. Consequently, many private and public sector entities have moved to provide as much of their information online as possible, as quickly as possible. Wikipedia, a collaborative encyclopedia, is an example of a private sector entity that provides all of its encyclopedia information online, as well as its internal information, such as quarterly reports. The U.S. Patent Office is an example of a public sector entity that provides all non-confidential proceedings online, as well as all internal data, such as employment related data and production statistics. The philosophy behind providing the aforementioned data online is that a better informed group of consumers will act in a more informed fashion, thereby increasing education and productivity, reducing costs and enhancing individuals’ experiences.
II
One area that has not evolved fast enough to keep up with this movement, however, is the web surfing or web browsing process. Today, it is still difficult and complex for regular web surfers to fully understand an online document that includes words or phrases that the reader does not understand. Typically, a reader faced with this situation will be forced to open a separate web browser window, engage in some interacting with the web browser to find the definition or explanation of the unrecognized word or phrase, and navigate back to the original web browser to understand the online document. This process can be tedious and annoying for web users and can lead to a disjointed and unsatisfying reading experience.
III
Therefore, a dire need exists for improvements over the prior art, and more particularly, there is a need for a more efficient method and system for facilitating the use and comprehension of web documents for consumers.
Summary
According to the aspects illustrated herein, a method on a server for providing annotation data for a web document over a communications network is disclosed. The method includes:
a) receiving, via the communications network, metadata about keywords, wherein the receiving step is performed continuously in real-time; b) generating annotation data for each keyword based on the metadata for each keyword; c) storing and indexing the annotation data in a connected database; d) providing, via the communications network, a graphical user interface for a user to peruse the web document, which includes a plurality of keywords; e) receiving, via the communications network, a message from the graphical user interface indicating that the user desires annotation data for a particular keyword of the plurality of keywords in the web document; f) performing a search for annotation data corresponding to the particular keyword in the connected database; and g) providing, via the communications network, the annotation data corresponding to the particular keyword, wherein said annotation data is displayed in the graphical user interface.
