JP2005196711A - Conversion of table information for rendering on area-restricted display - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an improved system for converting the table information including at least one table, its related terminal, method and computer program product. <P>SOLUTION: This system for converting the table information to be rendered to the terminal having an area-restricted display, comprises a network entity. The network entity includes a document converter capable of receiving an electronic document including the table information composed of the table. The document converter can convert the table into the straightened display of the table on the basis of the order of each table and on the basis of at least one label when at least one table includes at least one label. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates generally to a system and method for rendering table information, and more particularly to a system and method for storing memory for converting table information for rendering to a terminal with a limited display area. And computer program products.

  Companies and individuals who create electronic documents such as web pages, e-mail messages, and word process files typically have a target platform in mind. In many cases, the platform is a desktop computer with moderate storage capacity, memory, bandwidth and a large (usually at least 14 inches diagonal) display. Files created with this target platform in mind often contain complex formatting information with tables, frames, graphics, and navigation aids, all of which define how the document should be rendered. is there.

  A user may wish to access such a document via the medium on which the document was not originally configured, such as an Internet-enabled mobile phone, a personal digital assistant (PDA), or a handheld computer. These devices have limited display area, resolution, and rendering capabilities and cannot render the document as originally configured, or can only render in a very slow and / or inconvenient manner. For example, Internet-enabled mobile phones can usually display only a few lines of text, while their ability to render images is limited to grayscale or thumbnail size video or no video at all. . An extreme example of a medium that is limited in display is audio. In this regard, AT & T Natural Voices® “Text-to-Speech Engine” allows a user to dial an electronic document by dialing to a service that dictates the content of the document via the phone using text-to-speech conversion. Allow access. However, this dictation is of course a one-dimensional “rendering” of text and cannot represent complex layout information embedded in a two-dimensional table.

  Traditionally, devices that are unable to render a document as originally configured due to limitations in screen size, resolution, and rendering capabilities are free to render such a document when called. In this regard, for documents that contain information that is two-dimensionally formatted, such as tables, certain conventional devices eliminate such tables. Other devices cut one or more selected rows to fit within the display area of the device when the table is too wide. Yet another device performs a mechanical “linearization” of the table, typically ignoring the natural ordering of the contents of the cells in the table, eg from left to right, top to bottom. Each cell of the table is rendered independently while visiting the cells in the table in a predetermined order. In addition, other devices render only a partial “viewport” of the entire table and the user of the device wishes to move the viewport in the table, for example from left to right, thus exposing a different part of the table. (E.g., by pressing an arrow key). Such a technique for rendering a table is sometimes sufficient, but it provides an area-limited display so as to provide all table data and to make the information in the table easier to understand for the user viewing the information. It would be desirable to construct an improved system and method for converting table information for rendering into

  In view of the above background, embodiments of the present invention provide an improved system and associated terminals, methods and computer program products for converting table information including at least one table. In contrast to the prior art, embodiments of the present invention are convenient and understandable for the user viewing the information, for example by localizing the labels in the table into cells associated with each label. To provide the contents of the table, the table can be converted for later rendering by a terminal having a limited area display. Therefore, in contrast to conventional techniques for rendering a table to such a terminal, embodiments of the present invention do not cut one or more selected columns of such a table, Rather than performing a mechanical linearization of the table by rendering each cell of the table ignoring the natural order of the contents of the cells in the table, and rendering only a partial viewport of the entire table, Rather than requiring the user of the device to move the viewport, the entire contents of such a table can be rendered.

  In accordance with one aspect of the present invention, a system for converting table information for rendering to a terminal having a limited area display is provided. The system comprises network entities such as a mobile terminal, origin server, proxy / gateway, short message service center (SMSC), multimedia message service center (MMSC), gateway GPRS service node (GGSN), user processor and the like. The network entity comprises a document converter capable of receiving an electronic document containing table information consisting of table (s). In this regard, the document converter may determine the table line (s) based on the order of each table and also based on at least one label when at least one table includes at least one label. Can be converted into a customized display. More specifically, the document converter may convert the tables by configuring the table (s) so that each table has a predetermined order, such as a row-oriented order or a column-oriented order. it can. The document converter then localizes each label when the table includes at least one label, and then converts the table into a table configured to have a predetermined order, and the table further includes at least one label. When included, it can be linearized based on each localized label.

  The document converter can determine whether the terminal supports panning. If the terminal does not support panning, the document converter can receive a choice of either converting the table or transferring the electronic document to the terminal or the like without converting the table. Then, when the selection includes table conversion, the document converter can convert the table.

  As will be apparent, each table can and usually includes a plurality of cells organized in at least one row and at least one column. Therefore, the document converter constructs a table by identifying the natural order of the table (s), and when the table has a natural order other than a predetermined order, A table having an order can be configured such that each table has a predetermined order. The document converter can determine the natural order in a number of different ways. For example, the document converter can determine a characteristic vector for each cell in the table and determine a distance between adjacent cells based on the characteristic vector for each cell. The document converter can then determine the average distance between adjacent cell pairs in the same row of the table and the average distance between adjacent cell pairs in the same column of the table. The document converter can then identify the natural order of the table based on these average distances.

  The document converter can similarly localize the label (s) in each table in a number of different ways. For example, the document converter can identify one or more span labels and / or one or more direct labels. As will be apparent, each span label is associated with multiple columns or rows of the table, and each direct label is associated with a single column or row of the table. After identifying the span label or direct label, the document converter determines a label string for each cell of the table associated with the span label and / or direct label, thereby each span label for the table cell to which the label points. And each direct label can be localized, and each label string includes at least one of a span label and a direct label.

  The document converter can identify the direct label (s) based on a measure of similarity between adjacent rows and adjacent columns of the table. For example, the document converter can determine a measure of similarity between each pair of adjacent rows or columns. For example, the document converter determines the characteristic vector for each cell in each pair of adjacent rows or columns, and determines the distance between cells in the same column or row based on the characteristic vector for each cell, respectively. A measure of gender can be determined. The document converter can then add the distances across adjacent row columns or adjacent column rows to determine a measure of similarity between adjacent rows or columns, respectively. Then, after determining the similarity measure, the document converter can identify adjacent row or column pairs that differ by a given percentage or more from other pairs of adjacent rows or columns based on the similarity measure. . The document converter can then identify each identified pair of rows or columns of adjacent rows or columns as including at least one direct label.

  According to another aspect of the present invention, a terminal, method and computer program product for converting table information comprising at least one table are provided. Therefore, embodiments of the present invention provide a system and associated terminal, method and computer program product for converting table information comprising at least one table. The system and associated terminals, methods, and computer program products of embodiments of the present invention can convert tables for subsequent rendering by terminals such as mobile phones. In contrast to the prior art, the system and associated terminals, methods and computer program products can convert tables so that the terminal can provide the full contents of the table. In addition, the system and associated terminals, methods, and computer program products can localize the labels in the table to the cell associated with each label so that the table can be rendered in a manner consistent with the initial configuration of the table. . Therefore, the system and related terminals, methods, and computer program products of the embodiments of the present invention solve the problems shown by the known technology and exert additional effects.

Having generally described the invention, the invention will now be described in detail with reference to the accompanying drawings.
The present invention will now be described in detail with reference to the accompanying drawings, in which preferred embodiments of the invention are shown. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments will fully convey the concept of the invention to those skilled in the art. It is described to convey to. Like elements are designated by like reference numerals throughout.

  FIG. 1 illustrates one type of terminal and system that would benefit from the present invention. The systems, terminals, methods and computer program products of embodiments of the present invention will be described primarily in the context of mobile communication applications. However, it should also be understood that the systems, terminals, methods and computer program products of embodiments of the present invention can be used in connection with various other applications in the mobile communications industry and other industries. For example, the systems, terminals, methods, and computer program products of embodiments of the present invention can be used in connection with wireline and / or wireless network (eg, Internet) applications.

  As shown, the terminal 10 includes an antenna 12 for transmitting signals to and receiving signals from a base site or base station (BS) 14. This base station is part of a cellular network that includes the elements necessary to operate the network, such as a mobile switching center (MSC) 16. As will be apparent to those skilled in the art, a cellular network is also referred to as a base station / MSC / interworking function (BMI). In operation, the MSC can route calls to and from the terminal as the terminal places and receives calls. The MSC can also provide a connection to the landline trunk when the terminal joins the call. In addition, the MSC can control the transfer of messages to and from the terminal, as well as the transfer of terminal messages to and from the message center, eg, short message service to and from the SMS Center (SMSC) 17. The transfer of (SMS) messages can also be controlled.

  The MSC 16 can be connected to a data network such as a local area network (LAN), a metropolitan area network (MAN) and / or a wide area network (WAN). The MSC can be directly connected to the data network. However, in one exemplary embodiment, the MSC is connected to a proxy 18, a gateway (GTW), etc., and the proxy is connected to a WAN, such as the Internet 20. For example, the MSC can connect to a Wireless Application Protocol (WAP) GTW, which is connected to the Internet. Devices such as processing elements (eg, personal computers, server computers, etc.) can then be connected to the terminal 10 via the Internet. For example, as described below, the processing elements can include one or more processing elements associated with origin server 22 or the like, one of which is shown in FIG.

  The BS 14 can also be connected to a signaling GPRS (General Packet Radio Service) support node (SGSN) 24. As known to those skilled in the art, an SGSN can typically perform functions similar to MSC 16 for packet switched services. The SGSN can be connected to a data network such as the Internet 20 like the MSC. The SGSN can be directly connected to the data network. However, in a more general embodiment, the SGSN is connected to a packet switched core network such as GPRS core network 26. The packet-switched core network is then connected to another GTW, for example a GTW GPRS support node (GGSN) 28, which is connected to the Internet. The GGSN can also be connected to a message center, such as a multimedia message service (MMS) center 30. In this regard, the GGSN and SGSN can control the forwarding of messages such as MMS messages, similar to the MSC. The GGSN and SGSN can also control the transfer of terminal messages to and from the message center.

  Further, by connecting the SGSN 24 to the GPRS core network 26 and the GGSN 28, a device such as the origin server 22 can be connected to the terminal 10 via the Internet 20, SGSN and GGSN. In this regard, a device such as an origin server can communicate with the terminal across SGSN, GPRS, and GGSN. For example, the origin server can supply content to the terminal based on a multimedia broadcast multicast service (MBMS). For detailed information on MBMS, please refer to 3GPP TS 22.146, 3rd Generation Partnership Project (3GPP) Technical Specification entitled “Multimedia Broadcast Multicast Service (MBMS)”, the contents of which are hereby incorporated by reference.

  Terminal 10 may further connect to one or more wireless access points (APs) 32. An AP is an access point configured to communicate with a terminal based on technologies such as, for example, radio frequency (RF), including WLAN technology, Bluetooth (BT), infrared (IrDA), or a number of different wireless network technologies. . In addition or alternatively, the terminal may be connected to one or more user processors 34. Each user processor may include a computer system such as a personal computer or a laptop computer. In this regard, the user processor can be configured to communicate with the terminal based on technologies such as RF, BT, IrDA or many different wireline or wireless communication technologies including LAN and / or WLAN technologies. In addition or alternatively, one or more of the user processors can include a removable memory that can store the content, which can then be transferred to the terminal.

  The AP 32 and the user processor 34 can be connected to the Internet 20. Similar to the MSC 16, the AP and user processor can be directly connected to the Internet. However, in one effective embodiment, the AP is indirectly connected to the Internet via the proxy 18. As will be apparent, by connecting the terminal and origin server 22 and any of a number of other devices directly or indirectly to the Internet, the terminals communicate with each other and with the origin server, etc. Various functions can be performed, for example, data, content, etc. can be sent to and / or received from the origin server. As used herein, “data”, “content”, “information” and like terms are used interchangeably to refer to data that can be transmitted, received and / or stored in accordance with embodiments of the present invention. be able to. Accordingly, the use of such terms does not limit the spirit and scope of the present invention.

  FIG. 2 illustrates a block diagram of a network entity that can operate as a network entity, eg, terminal 10, origin server 22, proxy 18, SMSC 17, MMSC 30, GGSN 28, and / or user processor 34, according to an embodiment of the present invention. Based on. Although shown as separate entities, in some embodiments, one or more of one or more of a terminal, proxy, origin server and / or user processor that are logically separated but co-located within the entity May be supported. For example, logically separate but co-located terminals and proxies can be supported by a single entity. It is also possible, for example, to support logically separate but co-located origin servers and user processors with a single entity.

  As shown, the network entity may generally include a processor 36 connected to memory 38. The processor can also be connected to at least one interface 40 and other means for transmitting and / or receiving data, content, etc. The memory is a volatile and / or nonvolatile memory, and usually stores content, data, and the like. For example, the memory typically stores software applications, instructions, etc. for the processor to perform steps related to the operation of the entity in accordance with embodiments of the present invention. Further, as described below, the memory can also store a document converter that can convert table information for rendering by the terminal 10. For example, the memory may also store content that is transmitted from or received by a network entity, such as another network entity.

  FIG. 3 is a functional diagram of a mobile station that can operate as a terminal 10 according to an embodiment of the present invention. It should be understood that the mobile station shown and described below is merely illustrative of one type of terminal that would benefit from the present invention and therefore does not limit the scope of the invention in any way. . Many embodiments of mobile stations are illustrated and described below, but other types of mobile stations such as portable digital assistants (PDAs), pagers, laptop computers, and other types of voice and text communication systems are also available. The present invention can be easily used.

  The mobile station includes a transmitter 42, a receiver 44, and a processor such as a controller 46 that provides signals to and receives signals from these transmitters and receivers. These signals also include signaling information based on applicable cellular system air interface standards and user speech and / or user generated data. In this regard, the mobile station can operate with one or more air interface standards, communication protocols, modulation types, and access types. More specifically, the mobile station may operate based on any of a number of first generation (1G), second generation (2G), 2.5G and / or third generation (3G) communication protocols, etc. it can. For example, the mobile station can operate based on 2G wireless communication protocols IS-136 (TDMA), GSM and IS-95 (CDMA). Also, for example, the mobile station can operate based on 2.5G wireless communication protocol GPRS, Enhanced Data GSM Environment (EDGE), etc. In addition or alternatively, the mobile station may operate based on any of a number of different digital broadcasting technologies such as DVB technology (eg, DVB-T, ETSI standard EN300 744). A mobile station may also operate based on any of a number of different broadcast and / or multicast technologies such as MBMS technology (eg, 3GPP TS 22.146). Furthermore, the mobile station can also operate based on ISDB-T, DAB, ATSC technology, etc. Certain narrowband AMPS (NAMPS) as well as TACS mobile stations can benefit from embodiments of the present invention as well as dual or higher mode mobile stations (eg, digital / analog or TDMA / CDMA analog telephones). Can do.

  The controller 46 includes the circuitry necessary to perform the voice and logic functions of the mobile station. For example, the controller comprises a digital signal processor device, a microprocessor device, and various analog / digital converters, digital / analog converters and other support circuits. Mobile station control and signal processing functions are allocated between these devices based on their respective capabilities. Thus, the controller also includes the ability to convolutionally encode and interleave messages and data prior to modulation and transmission. The controller may further include an internal voice coder (VC) 46A and may include an internal data modem (DM) 46B. Further, the controller may include functionality for operating one or more software applications that can be stored in memory.

  The mobile station also includes a user interface including a conventional earphone or speaker 48, a ringer 50, a microphone 52, a display 54, and a user input interface, all connected to the controller 46. The user input interface that allows the mobile station to receive data is any of a number of devices that allow the mobile station to input data, such as a keypad 56, a touch display (not shown), or other input device. Can be included. In embodiments that include a keypad, the keypad comprises conventional numeric keys (0-9) and associated keys (#, *) and other keys used to operate the mobile station.

  The mobile station can also obtain and / or obtain data from electronic devices such as another terminal 10, proxy 18, origin server 22, AP 32, user processor 24, etc., based on any of a number of different wirelines and / or wireless technologies. Or it may include one or more means for sharing it. For example, the mobile station can include a radio frequency (RF) transceiver 58 and / or an infrared (IR) transceiver 60 so that it can acquire and / or share data based on radio frequency and / or infrared technology. Also, for example, the mobile station can include a Bluetooth (BT) transceiver 62 so that it can obtain and / or share data based on Bluetooth transfer technology. Although not shown, the mobile station may additionally or alternatively send data to the electronic device based on a number of different wireline and / or wireless network technologies, including LAN and / or WLAN technologies, and / or Data can also be received from there.

  In addition, the mobile station can typically include a memory, such as a subscriber identity module (SIM) 64, a removable user identity module (R-UIM), which stores information elements associated with the mobile subscriber. In addition to the SIM, the mobile station can include other memories. In this regard, the mobile station can include volatile memory 66 and / or other non-volatile memory 68, which can be embedded and / or removable. For example, other non-volatile memories can include embedded or removable multimedia memory cards (MMC), memory sticks, EEPROM, flash memory, hard disks, and the like.

  Memories 64, 66, 68 may also store multiple pieces of information and data used by the mobile station to perform the functions of the mobile station. For example, the memory may include an International Mobile Equipment Identification (IMEI) code, an International Mobile Subscriber Identification (IMSI) code, a Mobile Station Services Integrated Digital Network (MSISDN) code, etc. that can uniquely identify the mobile station to the MSC 16. An identifier can be stored. The memory may also store content such as received from origin server 22 and / or user processor 34. Also, for example, the memory can store one or more presentation applications such as a conventional text viewer, audio player, video player, multimedia viewer, and the like. Further, as described below, the memory can also store a document converter that can convert table information for rendering by the mobile station.

  As described in the background art, terminal 10 is generally unable to render an electronic document as it was originally configured due to limitations in terminal display area, resolution, and rendering capabilities. For example, many terminals can usually render on a display (eg, display 54) with only a few lines of text and a grayscale or thumbnail size image or no image at all. In this regard, the terminal is generally unable to render the table information of the electronic document as it was originally configured. For example, some conventional devices delete such table information. Other devices either excise the selected column (s) of the table or perform a mechanical linearization of the table, usually ignoring the natural order of the contents of the cells in the table Render each table cell independently. Yet another device renders only a partial viewport of the entire table, requiring the user of the device to move the viewport in the table, for example from left to right, thus exposing a different portion of the table. .

  Thus, embodiments of the present invention can convert an electronic document for rendering by a terminal having a display with limited display area, resolution, and / or rendering capabilities. More particularly, embodiments of the present invention can convert table information of an electronic document for rendering by such a terminal. Thus, in contrast to conventional techniques for rendering a table in such a terminal, embodiments of the present invention do not cut one or more selected columns of such a table, but instead in the table. Rather than performing a mechanical linearization of the table by rendering each cell of the table ignoring the natural order of the contents of the cells, and rendering only a partial viewport of the entire table, Rather than requiring the user to move the viewport, the entire contents of such a table can be rendered.

  According to an embodiment of the present invention, the table information of an electronic document generally comprises one or more tables each encapsulating a two-dimensional matrix of content such as data, information, etc. As will be apparent to those skilled in the art, such a two-dimensional table typically includes at least one row and at least one column. Also, such a table may include a complex structure (eg, a nested table or video), but usually does not. In addition, there is typically a high level of both syntactic and semantic coherence between the row (s) and column (s) of such a table.

  4 and 5 will be described. These are respectively a functional block diagram of the proxy 18 and a flowchart of a method for converting table information of an electronic document according to an embodiment of the present invention. More specifically, FIG. 4 illustrates an electronic document received from a content source 100 such as terminal 10, origin server 22, SMSC 17, MMSC 30, GGSN 28, user processor 34, and the like, and thereafter, according to one advantageous embodiment of the present invention. It is a functional block diagram of the proxy forwarded to. However, prior to transferring the electronic document, the proxy can operate the document converter 102, which receives the electronic document and then converts the table information of the document for rendering by the terminal 10. Can do. The proxy then transfers the electronic document containing the converted table information to the terminal and renders it on the terminal, or more specifically on the terminal display (eg, display 56).

  The proxy 18 operates the document converter 102 shown and described, but it will be understood that the document converter can operate from any of a number of different network entities in the system including, for example, the terminal 10 itself or the content source 100. I want. In this regard, the document converter can be implemented in a single network entity, or each part of the document converter can be implemented in more than one network entity. In addition, as described herein, the document converter is usually composed of software that can be stored in a memory (eg, memory 38) and operated by a processor (eg, processor 36). However, it should be understood that the document converter may be configured in firmware or hardware without departing from the spirit and scope of the present invention. Further, it should be understood that the document converter can perform operations on the electronic document other than table conversion to facilitate rendering of the electronic document by the terminal. Detailed information on various such operations can be found, for example, in US patent application Ser. No. 09/09, filed “Reorganizing Content of an Electronic Document” filed May 8, 2001, the contents of which are incorporated herein by reference. No. 851,404, published March 6, 2003 as US Patent Application Publication No. 2003/0046318.

  As shown in block 104 of FIG. 5, a method for converting table information including at least one table comprises a document converter 102 that receives an electronic document from a content source 100. According to an embodiment of the present invention, an electronic document includes one or more tables, and each table includes at least one row of content and at least one column of content. Electronic documents can take any of a number of different formats. For example, an electronic document may be a hypertext transfer protocol (HTTP) document, an e-mail document, a portable document format (PDF) document, a postscript document, an ASCII text format (TXT) document, an extensible markup language (XML) document. , Microsoft Word (registered trademark) (DOC) documents, Microsoft Excel (registered trademark) (XLS) documents, and the like.

  After receiving the electronic document, document converter 102 may convert the electronic document into an intermediate data structure representation of the electronic document, as shown at block 106. The document converter can convert an electronic document in any of a number of different ways to generate any of a number of different intermediate data structures that can represent the electronic document. In this regard, the document converter can convert an electronic document into an intermediate data structure consisting of a common internal tree-based representation of the electronic document. For example, a document converter can convert an electronic document into an extensible hypertext markup language (XHTML) document object model (DOM) representation of the electronic document.

  Document converter 102 can convert an electronic document into an XHTML DOM data structure in a number of different ways. In one embodiment, for example, the data converter converts the electronic document to an HTML document (if not already an HTML document), then converts the HTML document to an XHTML document, and then converts the XHTML document to an XHTML DOM data structure. To convert the electronic document. In order to convert an electronic document to a document of the above format and generate an XHTML DOM data structure, the document converter can execute any of a number of different known conversion routines.

  In addition or alternatively, the document converter 102 can communicate with another software, hardware or firmware module running on the proxy 18 or another network entity, where the other module is a document One or more conversion routines can be executed to help the converter convert the electronic document to an XHTML DOM data structure. For example, a document converter can communicate with a wvWare software package developed under the wvWare project to convert a DOC document into an HTML document. Also, for example, a document converter can communicate with an xlHTML software package developed under the Chicago project to convert an XML document to an HTML document. Then, for example, the document converter can communicate with an HTML Tidy software package developed under the Tidy project to convert the HTML document to an XHTML document.

Further, for example, the document converter 102 may communicate with an XML parser, such as the Xerces XML Parse software package (developed under the Apache XML project), to convert an XHTML document into an XHTML DOM data structure. it can. Regardless of how the document converter converts an electronic document into a data structure that represents the electronic document, the data structure is stored in volatile memory, not in the non-volatile memory in which the document converter and electronic document normally reside. You can remember, but it doesn't have to be. As will be apparent, the data structure typically includes a number of nodes corresponding to the tags of the transformed XHTML document based on the tags of the underlying HTML document. For example, consider the following source code for an HTML document.
<html>
<body>
<img src = “http://www.domain.com/img/image.gif” />
<table>
<tr>
<td> This is the first cell </ td>
<td><img src = ”http://www.domain.com/img/image2.gif” /></td>
<td> This is the third cell </ td>
</ tr>
<tr>
<td> This is the fourth cell </ td>
<td><img src = ”http://www.domain.com/img/image3.gif” /></td>
<td> This is the sixth cell </ td>
</ tr>
</ table>
<small>
A text block
</ small>
</ body>
</ html>

  As shown in the source code above, the HTML document includes a 2 × 3 table, and the second column of each row includes an image. The HTML document also includes a third image and a small font block of text. From such an HTML document, the document converter can generate an XHTML DOM data structure as shown in FIG.

  Regardless of how the document converter 102 converts an electronic document into an intermediate data structure representation of the electronic document, when the document converter converts or converts the electronic document into the intermediate data structure, the document converter As shown in block 108 of FIG. 5, an “implicit” table in the intermediate data structure may be identified. Obviously, in the format of an electronic document, for example, in the case of a table identified by an HTML document tag (ie, “<table> </ table>”), the document converter usually makes such a table especially There is no need to identify. However, in the case of other tables, the electronic document is configured such that the information is rendered in table form, but does not include a format configuration of information such as a table. The document converter can identify such an “implicit” table, where the implicit table is usually defined by the table node of the data structure after the document converter converts the electronic document into a data structure. Do not show. The document converter can then reformulate the data structure to include a table node for such a table.

For example, consider the following ASCII format text in a TXT document.
Activity Calories / hour Aerobics 660
Basketball 550
Running 925
Tennis 450
Visually, the ASCII format text clearly represents a table. However, the TXT format does not include a mechanism (eg, a tag) for explicitly indicating text such as a table. Based on the technique used by the document converter to convert such a TXT document into a data structure, the document converter recognizes such a text block as a table. However, in various other instances, the document converter 102 may not recognize such a text block as a table. In such instances, the document can generate a data structure as shown in FIG. 7A. As shown, this data structure identifies the text as a block of unstructured text. Thus, the document converter is implicit in instances where the electronic document includes an implicit table that is not recognized by the document converter during the conversion of the electronic document, or includes an implicit table represented as a table node in the data structure. The target table can also be identified.

The document converter 102 can identify the implicit table in any of a number of different known ways, including through a dedicated parser and automatic learning techniques. For a number of examples of known techniques for identifying implicit tables, see “A System of Understanding and Reformulating Tables, Fourth ICPR WORKSHOP ON DOCUMENT ANALYSIS SYSTEM” by Jianning Hugh et al. (2000); Mathew Hurst & Shona Douglas, Layout and Language: Preliminary Investigations in Recognizing the Structure of Tables, PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON DOCUMENT ANALYSIS AND RECOGNITION (ICDAR) 1043-1047 (1997); etc. "Learning to Recognize Tables in Free Text, Proceedings of the 37 th Annual Meeting of ACL 443-450 (1999) ", which is incorporated herein by reference.

  Regardless of how the document converter 102 identifies the implicit table, if the document converter identifies one or more implicit tables of the electronic document, the document converter will identify the text as a table. Can be generated or reformulated. For example, the document converter recognizes a block of text as a table, and converts the block of text to an appropriate HTML tag (eg, <table>, <tr>, <td> while converting the electronic document to an HTML document. Etc.), an HTML document formatted to identify a block of text as a table can be generated. Continuing the above example for a block of text in a TXT document, the document converter identifies the block of text as an implicit table and then includes the implicit table as an explicit table, as shown in FIG. 7B. A data structure can be generated or reformulated.

  After the document converter 102 has generated a data structure representation of the electronic document that includes reshaping the implicit table as an explicit table, the document converter has the electronic order to have a predetermined order, as shown at block 110. Document tables can be reconstructed. In this regard, the document converter can identify whether the natural order of cells in the table of electronic documents is row-major or column-major. As will be apparent to those skilled in the art, tables generally have row-major order when the contents of the table are composed of rows. Conversely, a table generally has a column-major order when the contents of the table are composed of columns.

  After identifying the natural order of the cells in the table of electronic documents, the document converter 102 can reconstruct the tables so that all tables have a predetermined order. In one exemplary embodiment, for example, the document converter can reconstruct the tables so that all tables have a row-based order. In such a case, for a table with column-major order, the document converter can transpose the table so that the natural order is row-dominant. However, it will be appreciated that the document converter can also reconstruct the table so that the table has a column-major order, in which case other operations of the document converter are adjusted accordingly.

  The document converter 102 can identify the natural order of the table in any of a number of different ways. For example, the document converter can be configured to search for a pattern of similarity among neighboring cells in each table. In this regard, for example, consider a column-based table having a second row that contains an image in each cell. The fact that all cells in such a row are similar to each other is usually not critical, but is evidence that the table has a column-major order. Separately, for example, consider a row-based table having a second column that contains an image in each cell, and a fifth column that contains a percentage sign in each column after a number. The similarity of column cells to each other is also evidence that the table has row-major order, although usually not definitive.

  In one exemplary embodiment, document converter 102 identifies the natural order of the table by representing each cell's information as a characteristic vector. For example, each cell can be represented by a characteristic binary vector of a given length N, where each position of the vector is the value of a predicate (yes / no question) that represents the characteristic of the information in each cell. Correspond. Obviously, the characteristic vector display is a method of encoding the structural characteristics of each cell of the table. Thus, cells that look similar usually have similar characteristic vector representations.

The characteristic vector can include a number of bits (N) corresponding to a number of different predicates. For example, the characteristic vector can include 13 bits (ie, N = 13), with each bit corresponding to a predicate as follows:
Bit 1 is “1” if and only if the cell contains an image.
Bit 2 is “1” if and only if the cell contains digits.
Bit 3 is “1” if and only if the cell contains a hyper ring.
Bit 4 is “1” if and only if the cell contains bold text.
Bit 5 is “1” if and only if the cell contains italic text.
Bit 6 is “1” if and only if the cell contains punctuation marks.
Bit 7 is “1” if and only if the cell contains 0 to 5 characters.
Bit 8 is “1” if and only if the cell contains 6 to 10 characters.
Bit 9 is “1” if and only if the cell contains 11 to 15 characters.
Bit 10 is “1” if and only if the cell contains 16 to 20 characters.
Bit 11 is “1” if and only if the cell contains 21 to 25 characters.
Bit 12 is “1” if and only if the cell contains 26 to 30 characters.
Bit 13 is “1” only when the cell includes 31 characters or more.

8A and 8B showing the table and the characteristic vector for the table cell, respectively, will be briefly described. As shown for such a table, for example, the cells in the first row, first column contain blanks. Thus, the cells in the first row, first column can be represented by the characteristic vector 0000001,000,000, where a unique “1” bit value corresponds to a cell containing 0 to 5 characters. In contrast, the second row, the cells of the first column includes the phrase "equity reserve ^SM line (equity reserve ^SM line)", which can be expressed by the feature vector 0000001000000. As shown, a unique “1” bit value corresponds to a cell containing 21 to 25 characters (this phrase includes a counting space of 21 characters).

  After representing the information for each cell as a characteristic vector, document converter 102 can determine a similarity measure between cells in the table based on the characteristic vector. The document converter can determine the value of similarity between adjacent cells in a number of different ways, for example, by determining the “Manhattan” distance between adjacent cells. In this regard, the distance can be determined for two neighboring cells by determining the number of bits that differ between the characteristic vectors of each cell. For example, the distance between 0111 and 1110 is 2. Obviously, the smaller the distance between adjacent cells, the greater the similarity of each cell. See FIG. 8B for distance measures between cells in the table of FIG. 8A determined in accordance with embodiments of the present invention.

  Also, as will be apparent, in various cases, one or more predicates corresponding to each bit value (s) of the characteristic vector are more strongly indicative of cell similarity than other predicates. For example, the first bit of the 13-bit characteristic vector described above represents the cell similarity more strongly than the tenth bit. Therefore, according to an embodiment of the present invention, one or more bits have an associated weight so that the distance between two characteristic vectors includes the associated weight as a sum when each bit is different between adjacent cells. Can be.

上記式（１）及び（２）に示すように、ｉ＝１・・・Ｒは、テーブルにおける行を表わし、そしてｊ＝１・・・Ｃは、列を表わす。又、ｘ_i,j及びｘ_i+1,jは、テーブルの隣接セルを表わし、そしてΔ_Cは、各隣接セル間の距離を表わす。図８Ａのテーブルについては、図８Ｂに示す距離から、Ｓ_H＝２．４（即ち、２４／１０）であり、そしてＳ_V＝０．５８３（即ち、７／１２）である。 Regardless of how exactly the similarity measure or distance between adjacent cells of the table is determined, after the distance is determined, the document converter 102 has either a row-major order or a table-major order. Can be determined. More specifically, the document converter can determine the order of the table based on the distance across various rows and columns of the table. For example, a document converter can determine the average distance S _H between adjacent cell pairs in the same row in the table, and determines the average distance S _V between adjacent cell pairs in the same column in the table. As is apparent, a small average distance between adjacent cells in the same row usually indicates that the rows of the table contain similar cells, which suggests that the table has a column-major order. Conversely, a small average distance between adjacent cells in the same column indicates that the table column contains similar cells, which suggests that the table has row-major order. When written in display, the document converter can determine S _H and S _V based on the following equations:

As shown in equations (1) and (2) above, i = 1... R represents a row in the table, and j = 1... C represents a column. X _{i, j} and x _{i + 1, j} represent adjacent cells in the table, and Δ _C represents the distance between each adjacent cell. For the table of FIG. 8A, from the distance shown in FIG. 8B, S _H = 2.4 (ie 24/10) and S _V = 0.583 (ie 7/12).

When the document converter 102 determines the average distance S _H between adjacent cell pairs in the same row and the average distance S _V between adjacent cell pairs in the same column, the document converter may determine whether the table has a row-oriented order or column-oriented. You can decide if you have an order. For example, the document converter can determine the order by comparing the average distances. If the average distance S _V between adjacent cell pairs in the same column is greater than the average distance S _H between adjacent cell pairs in the same row (ie, S _V > S _H ), the document converter has the table as column-oriented. It can be determined to have an order, otherwise the table has a row-oriented order. In order to reconstruct a table of an electronic document to have a predetermined order, such as a row-oriented order, the document converter converts a table having a column-oriented order so that all the tables have a row-oriented order. Can be transposed. In display terminology, the table of electronic documents can be transposed by reorganizing the DOM tree such that each cell x _{i, j} is exchanged for each cell x _{j, i} .

As will be apparent, in various cases, _SH and _SV are very similar, so that the table is transposed to have the opposite order as desired. Accordingly, the document converter 102, it is possible to weight the S _H or S _V in comparison, a document converter is slightly higher the likelihood of transposition a given table. For example, a document converter, the S _H, it is possible to multiply the bias C having a value greater than 1. Therefore, in comparison of S _H and S _V as _{(i.e., S V> C * S H} ), the bias is increased, the document converter transpose and S _H and S _V are similar it to column major table When it has the value, the likelihood of transposing the table is low.

  After restructuring the table to have a predetermined order, the document converter 102 can linearize the table into a one-dimensional sequence of the contents of the cells of the table, where the linearized content is the table. Not included. Conveniently, however, prior to linearizing the table, the document converter can localize the labels contained within the table, as shown in block 112 of FIG. As will be apparent, a number of tables contain cells that serve as labels for one or more other cells. In this regard, when a label refers to a single row or column of cells, the label is often referred to as a “direct” label. In contrast, when a label refers to more than one row or column, the label is often referred to as a “span” label. As shown in the table of FIG. 9, for example, labels such as “YR”, “TM”, “GP” and “G” constitute a direct label, while “Regular Season” and “Post Season” The label constitutes a span label.

  Also, as is apparent, when a row-based table having column labels as shown in FIG. 9 is linearized, the column labels can be separated from the cells pointed to by such labels. Therefore, it is difficult to understand the rendered table in the terminal 10 having a limited display area. In this regard, refer to FIGS. 10A and 10B, which show a portion of the table of FIG. 9 on a terminal display and the table is advanced by a simple linearization process. As is apparent, the row label is separated from the cell to which such a label points, thus making it difficult to understand the table. More specifically, as shown, the column containing a number of direct labels (eg, “YR”, “TM”, “GP”, “G”, etc.) is the data that it points to, ie under each label. Are visually separated from the cells in each column.

  Thus, in order to avoid separating column labels from the data pointed to by such labels, the document converter 102 of one effective embodiment uses labels (if any) placed in a table of electronic documents. ). More specifically, the document converter reconstructs the label so that the label is placed near the data that the label describes. Thus, the document converter can help the user understand the tables that are subsequently rendered.

  The document converter 102 can localize the labels in the table of electronic documents in any of a number of different ways. One particularly effective technique for localizing labels is described below with reference to FIG. As described below, the document converter localizes labels for row-based tables having at least one row that includes at least one label. However, it should be understood that the document converter can localize labels in a number of other situations. For example, the document converter can localize labels for a column-based table having at least one column that includes at least one label. In such cases, the process of localizing labels usually proceeds as described here, with operations performed based on rows being performed based on columns and vice versa. Can do. It should also be understood that the document converter need not localize the labels in the table, especially for tables that do not contain labels. Normally, a document converter does not localize a label even when it reorganizes the table into a row-based table and the table does not include a work with labels.

  According to one technique, document converter 102 can localize a label by first identifying a row that contains a span label, ie, a label that spans two or more columns, as shown in block 118 of FIG. . More specifically, the document converter can determine whether a cell has a predetermined number of rows (eg, 3 rows) for a tag that identifies a cell containing table data (eg, <td>) or a table header (eg, <th>). Where the cell also has a multi-column attribute (eg, colspan). As will be apparent, such attributes typically represent cells that span two or more columns of the table. For example, in the table of FIG. 9, the first row includes two cells (ie, “regular season” and “post season”), each spanning two or more columns.

  After identifying rows that contain span labels, document converter 102 may identify rows that contain direct labels, ie, labels that each point to a single column, as shown in block 120. Similar to the above, the document converter can inspect a predetermined number of rows (for example, 3 rows) of cells against the direct label. The document converter can search for direct labels in any of a number of different ways. For example, in one embodiment, the document converter searches for a direct label based on a similarity measure (eg, distance) between adjacent rows, where the similarity measure between adjacent rows is determined as described above. Based on similarity measure between cells. In this regard, as will be apparent to those skilled in the art, direct column label rows often differ from the remaining rows in the table in a number of different ways. For example, the second row of FIG. 9 consisting of direct column labels is composed entirely of letters, while the line below the second row is composed almost entirely of numbers.

この式において、ｍは、少なくとも１つのスパンラベルを含まないテーブルの第１行を表わす。 In order to search the direct labels of the table, the document converter 102 can determine the similarity or distance between adjacent rows of the table, where the similarity is the similarity of vertically adjacent cells in each row. To be determined. Written in notation, the document converter can determine the distance delta _R between adjacent rows of the table based on the following equation (3).

In this equation, m represents the first row of the table that does not include at least one span label.

After determining the distance between adjacent rows, document converter 102 can identify adjacent row pairs that have significantly greater differences than subsequent adjacent row pairs, where the significantly larger differences include direct labels. Represents the first row of each pair. In other words, the document converter determines whether the distance between a pair of adjacent rows is significantly greater than the distance between subsequent pairs of adjacent rows, and then the first row of such a pair includes a direct label. Identify as. More specifically, for example, the document converter can compare each Δ _R (i, i + 1) with the distance between subsequent pairs in adjacent rows, starting with Δ _R (i + 1, i + 2). Then, for a first Δ _R (i, i + 1) that is at least a given percentage (eg, 50%) greater than the subsequent pair of adjacent rows, row i can be identified as the row containing the direct label. .

  When document converter 12 identifies a line containing span labels and / or direct labels, it may generate a label tree representation of the table based on any of a number of known techniques, as shown in block 122 of FIG. Can do. More specifically, for example, the document converter can generate another data structure for use in localizing labels, where the data structure is spanned and direct to individual cells of the table. Reflect column label relevance. This data structure, referred to as a “label tree”, can capture the structure of previously identified span and direct column labels. In this regard, FIG. 12 shows a label tree representation of the table of FIG. 9 according to one embodiment.

  The document converter 102 can localize the labels of the table after it generates a label tree representation of the table. As shown in block 124, the document converter can localize the labels by traversing the label tree and data structure display and determine the placement of the labels within the linearized display of the table. Thus, the table can then be linearized to include a label placed close to each table cell. As will be apparent, the label tree display and data structure can be traversed in any of a number of different ways. In one embodiment, for example, the document converter may include a cell with no content (ie, a blank cell) and a column ( Each cell of i = 1,... C) is inspected.

For each column of each row containing blanks and label cells, document converter label path x _k = from tree label the root node of the table to the leaf L branch _{{x 1, x 2, ···} x L} (See FIG. 12), where each leaf is associated with each column of the table. In addition, the document converter can define a label string s _ij , where the label string is initialized with a value of 0 for each cell. For each node in each label path (k = 1,... L) in the label tree display of the table, x _{k comprises the} span label and x _{k + 1 comprises} the first node under x _k In some cases, the document converter 102 can add a new line to the label string s _ij with a span label, another new line, and at least one character space (eg, a tab). On the other hand, if x _k comprises a direct label, the document converter can add the direct label to the label string s _ij with a separator such as a colon. Then, after examining the table cells and generating the label strings s _ij , the document converter can attach each label string to each cell x _ij during table linearization as described below.

For example, consider the table of FIG. 9 and the label tree display of FIG. Here, the table includes a number of spans and direct labels. In such a case, the examination of the table cell can begin with cell (3, 1) containing the value “84-85”. Following the root node of the label tree display to the leaf associated with the third row, the document converter 102 can define a label path x _k = {x ₁ = “YR”}. In addition, the document converter can define a label string s _3,1 , where the label string is initialized with a value of zero. Then x ₁ contains the direct label, so the document converter can add the direct label along with the colon to the label string s _ij so that s _3,1 = {“YR:”}. The same process can be repeated for cell (3, 2) so that s _3,2 = {“TM:”}. On the other hand, for cell (3, 3), the document converter can define the label path x _k = {x ₁ = “regular season”; x ₂ = “GP”}. After initializing the label string s _3,3 to zero, x ₁ contains the span label and x ₂ (ie, x _{i + 1} ) contains the first node under x ₁ so that the document converter A new line can be added to the label string s _3,3 with a span label, another new line and tab. Then, with respect to x _2, a document converter further, it is possible to add direct label and colon label string s _3,3. Therefore, the label string s _3,3 appears as follows:
s _3,3 = {\ n; "regular season"; \ n; \ t; "GP:"}
However, “\ n” and “\ t” represent new lines and tabs, respectively, when the label string is rendered by the terminal 10.

Referring again to FIG. 5, when the document converter 102 determines the label string s _ij and localizes the labels, it can linearize the tabs into a one-dimensional sequence of cells, as shown in block 114. it can. The document converter can linearize tabs in any of a number of different ways. In one embodiment, for example, the document converter can linearize the table by replacing the table with one or more individual paragraphs (eg, <p>), each paragraph being one of the first table. Includes the above cells. In this regard, especially when the document converter configures the table of electronic documents to have a row-oriented order, the document converter usually starts with the upper left cell and ends with the lower right cell, so that A table with an order can be traversed and the contents of each cell can be moved to each paragraph of the linearized display of the table.

In one embodiment, the document converter 102 can linearize the table by examining the data structure representation of the table. For example, consider the data structure display of an electronic document including the table shown in FIG. 13A. For each table node, the document converter can create a new temporary node (shown as node “X” in FIG. 13B) in the data structure at the same level as each table node. Then, for each data node (ie, <td>) of each row node (ie, <tr>) that typically represents a table cell, except for blanks and label cells, the document converter converts a paragraph node (eg, <p>). It can be attached as a child of node “X”. As shown in FIG. 13B, the document converter can then move the content node of each data node (ie, C1, C2, etc.) or, more specifically, the content x _{ij of} each cell down to each paragraph node. In FIG. 13B, only the first and second content nodes C1 and C2 are shown as being moved. In addition to moving each content node below each paragraph node, the document converter can attach each label string s _ij to the contents of each cell if each cell has a corresponding label string.

  After moving the content node of each data node of each row node under each paragraph node, the document converter 102 can delete the table node and its associated row node, data node and content node under the table node. . The document converter can delete the table node and its related nodes after moving the content node of each data node of each row node, but the document converter has moved or deleted all children (ie related nodes) of each node Each node can be deleted later. Thus, for example, the document converter can delete each table data node (ie, <td>) after moving each content node (ie, C1, C2, C3, etc.) below each paragraph node. Similarly, for example, the document converter can delete each table row node (ie, <tr>) after deleting each table data node (ie, <td>).

However, regardless of when the document converter 102 deletes the table node, after moving the content node of each data node of each row node below each paragraph node, the document converter removes the node “X” from the data structure. As shown in FIG. 13C, each node under node “X” can be moved up in the data structure display of the electronic document. Accordingly, as shown, each content node (ie, C1, C2, etc.) representing each cell x _ij (excluding blanks and label cells) of the table is included in each paragraph node of the data structure display of the electronic document. be able to. As will be apparent, in such a case, each cell of the first table containing the attached label string s _ij can be included under another paragraph node. However, the document converter can linearize the table so that each row of cells of the table is included under each paragraph node. As such an example, see FIG. 14 which shows a linearized display of the table of FIG. Here, one or more cells of the table contain an attached label string.

  The document converter 102 converts the data structure representation of the electronic document (including the linearized representation of the table) back into an electronic document, as shown in block 116 of FIG. 5, when this linearizes the table. . As will be apparent, the document converter can convert the data structure in a number of different ways, for example, the opposite of how the document converter converted the electronic document to an intermediate data structure (see block 106). Can be converted. Then, when the document converter 102 converts the data structure back into an electronic document, the electronic document now includes a linearized representation of each table contained in the original electronic document, and the proxy 18 As indicated at 117, the electronic document can be transferred to the terminal 10. The terminal can then render the electronic document on a display (eg, display 54). In this regard, see FIG. 15 for an example of rendering the table of FIG. 9 on such a display.

  As shown and described, the document converter 102 can receive an electronic document that includes at least one table and then convert the table information of the document for rendering by the terminal 10. The document converter can then transfer the electronic document containing the converted table information to the terminal for rendering by the terminal or, more specifically, the terminal display (eg, display 56). Note that in various cases, the document converter can transfer an electronic document without converting the table information of the document. In this regard, as will be apparent, in various cases, the terminal can support left and right pans within a display window that is wider than the terminal display. Thus, the document converter can determine whether the terminal recipient of the electronic document supports left and right panning before converting the table information. The document converter can then transfer the electronic document without converting the table information. Alternatively, the document converter can request the terminal user to select whether to convert the table information of the electronic document or to transfer the electronic document without converting the table information.

  The document converter 102 can determine the terminal 10's ability to pan an electronic document in a number of different ways. For example, the document converter may maintain an internal “device capability” lookup table, which includes a list of one or more recipient devices, each indicating whether to support panning of electronic documents. . In such a case, the document converter determines the terminal recipient while establishing a communication session between the terminal and the content source 100 (eg, HTTP GET), and then in a “device capabilities” lookup table. A terminal lookup can be performed. Then, after requesting a selection of whether to convert the table information, the document converter can receive the selection and process it accordingly. In this regard, if the user chooses to transfer the electronic document without converting the table information, the document converter can operate appropriately. Otherwise, if the user chooses to convert the table information or if the terminal does not support panning, the document converter can convert the table information in the manner described here.

  Also, as shown and described, document converter 102 can be from any of a number of different network entities such as terminal 10, proxy 18, or content source, eg, origin server 22, SMSC 17, MMSC 30, GGSN 28, user processor 34, etc. But it can work. However, as will be appreciated, various operations of the document converter, particularly when converting table information comprising at least one table in an electronic document, can be performed independently of connection to the network. . Thus, it should be understood that the network entity operating the document converter is an entity such as a terminal or content source (eg, origin server, SMSC, MMSC, user processor, etc.) that is not connected to the network. In such a case, for example, the network entity can be provided with an electronic document in any of a number of different ways independent of the network, after which the document converter provides a table of such electronic documents as described above. Can be converted in a way.

  In accordance with one aspect of the present invention, all or part of the system of the present invention, eg, all or part of the terminal 10, proxy 18, origin server 22, SMSC 17, MMSC 30, GGSN 28 and / or user processor 34, in general, Operates under the control of a computer program product (eg, document converter 102). A computer program product for performing the method of an embodiment of the present invention includes a computer readable storage medium such as a non-volatile storage medium and a series of computer instructions implemented on the computer readable storage medium. Contains computer readable program code portions.

  In this regard, FIGS. 5 and 11 are flowcharts of methods, systems and program products according to the present invention. It will be understood that each block or step of the flowchart, and combinations of blocks in the flowchart, can be implemented by computer program instructions. These computer program instructions are loaded into a computer or other programmable device to generate a machine and the instructions executed on the computer or other programmable device are functions specified in the blocks or steps of the flowchart. Forming means for implementing These computer program instructions may also be stored in a computer readable memory to direct the computer or other programmable device to function in a particular manner and stored in the computer readable memory. The instructions form a manufacturing article that includes instruction means for performing the functions specified in the blocks or steps of the flowchart. Computer program instructions may also be loaded into a computer or other programmable device to cause a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process. The instructions executed on the programmable device provide steps for performing the functions specified in the blocks or steps of the flowchart.

  Thus, a block or step of the flowchart supports a combination of means for performing a specified function, a combination of steps for performing a specified function, and a program instruction means for performing the specified function. Also, each block or step of the flowchart, and combinations of blocks or steps in the flowchart, are implemented by a special purpose hardware-based computer system that performs a specified function or step, or a combination of special purpose hardware and computer instructions. It will be understood that it can be done.

  Numerous modifications and other embodiments of the invention will become apparent to those skilled in the art who have the benefit of the teachings set forth in the foregoing description and accompanying drawings. For example, as described above in one embodiment, the document converter configures or reconfigures the table such that the table has a row-major order, and then processes the table based on such row-major order. . However, in another embodiment, the document converter may configure or reconfigure the table such that the table has a column-major order, and adjust other operations of the document converter accordingly. For example, a document converter localizes labels for a column-based table having at least one column containing at least one label, and this localization process typically performs operations based on columns based on rows. It can be performed in the same way as described above, with execution and vice versa. Therefore, it is to be understood that the invention is not limited to the specific embodiments described above, but includes other embodiments within the scope of the claims. Also, although specific terms have been used, they are commonly used for purposes of explanation and are not intended to be limiting.

1 is a block circuit diagram of a wireless communication system according to an embodiment of the present invention including a cellular network and a data network with terminals connected in both directions via a wireless RF link. FIG. FIG. 4 is a block circuit diagram of an entity that can operate as a terminal, origin server, proxy / gateway, SMSC, MMSC, GGSN, and / or user processor in accordance with an embodiment of the present invention. FIG. 2 is a block circuit diagram of a mobile station that can operate as a terminal according to an embodiment of the present invention. FIG. 2 is a functional block diagram of a terminal that downloads or otherwise receives an electronic document that includes a table in accordance with one embodiment of the present invention. 6 is a flowchart illustrating various steps of a method for converting table information of an electronic document according to an embodiment of the present invention. 2 is a block diagram illustrating a tree-based data structure of an electronic document including a table according to an embodiment of the present invention. FIG. FIG. 3 is a block diagram illustrating a tree-based data structure of an electronic document including an implicit table according to an embodiment of the present invention. FIG. 7B is a block diagram illustrating the tree-based data structure of FIG. 7A with an implicit table reformulated as an explicit table in accordance with one embodiment of the present invention. It is a figure which shows the table contained in an electronic document based on one Embodiment of this invention. FIG. 8B is a block diagram of the table of FIG. 8A including a characteristic vector for the cells of the table and a similarity measure between the cells according to one embodiment of the present invention. It is a figure which shows another table contained in an electronic document based on one Embodiment of this invention. FIG. 10 is a schematic block diagram illustrating a portion of the table of FIG. 9 rendered on a terminal display, showing the table linearized without localizing the table labels. FIG. 10 is a schematic block diagram illustrating a portion of the table of FIG. 9 rendered on a terminal display, showing the table linearized without localizing the table labels. 6 is a flowchart illustrating various steps of a method for localizing labels for a table included in an electronic document according to an embodiment of the present invention. FIG. 3 is a block diagram showing a label tree of a table included in an electronic document according to an embodiment of the present invention. FIG. 3 is a block diagram illustrating a tree-based data structure of an electronic document that includes a linearized table according to one embodiment of the invention. FIG. 3 is a block diagram illustrating a tree-based data structure of an electronic document that includes a linearized table according to one embodiment of the invention. FIG. 3 is a block diagram illustrating a tree-based data structure of an electronic document that includes a linearized table according to one embodiment of the invention. FIG. 10 is a diagram showing a linearized table of FIG. 9 according to an embodiment of the present invention. FIG. 10 is a schematic block diagram illustrating a portion of the table of FIG. 9 rendered on a terminal display, showing the table converted according to one embodiment of the present invention.

Claims (40)

In a system that converts table information for rendering to a terminal with a limited area display,
Comprising a network entity comprised of a document converter capable of receiving an electronic document comprising table information comprising at least one table, said document converter comprising at least one table based on the order of each table and at least one table A system adapted to convert a linear representation of at least one table based on at least one label when the table includes at least one label.   The document converter configures at least one table such that the at least one table has a predetermined order, localizes each label when the at least one table includes at least one label, and then at least one By linearizing the table based on at least one table configured to have the predetermined order and also based on each localized label when the at least one table includes at least one label, The system of claim 1, wherein at least one table can be converted.   Each table includes a plurality of cells organized in at least one row and at least one column, wherein the document converter constitutes at least one table by identifying the natural order of the at least one table, and When at least one table has a natural order other than a predetermined order, at least one table having a natural order other than the predetermined order may be configured such that each at least one table has a predetermined order. The system according to claim 2, which is capable.   The document converter can determine a characteristic vector for each cell of the table, and determine a distance between adjacent cells based on the characteristic vector for each cell, and the document converter can also determine the same row of the table. The average distance between adjacent cell pairs in and the average distance between adjacent cell pairs in the same column of the table can be determined, and then the natural order of the table can be identified based on these average distances Item 4. The system according to Item 3.   3. The system of claim 2, wherein the document converter can configure at least one table such that the at least one table has one of a row-based order and a column-based order.   The document converter can determine whether the terminal supports panning and, if the terminal supports panning, either converts at least one table or does not convert the electronic document at least one table. The method of claim 1, further comprising receiving a selection of one to transfer to, and wherein the document converter is capable of converting at least one table when the selection includes conversion of at least one table. system. A system for localizing at least one label of a table in an electronic document, the table including a plurality of cells organized in at least one row and at least one column,
Comprising a network entity comprised of document converters capable of identifying at least one of at least one span label and at least one direct label, each span label being associated with one of a plurality of columns and rows of the table; and Each direct label is associated with one of a single column and a single row of the table, and the document converter determines a label string for each cell of the table associated with at least one of a span label and a direct label. Thus, each span label and each direct label can be localized, and each label string includes at least one of a span label and a direct label.   8. The system of claim 7, wherein the document converter is capable of identifying at least one direct label based on a measure of similarity between one of adjacent rows and adjacent columns of a table.   The document converter determines a measure of similarity between each pair of adjacent rows and one of each pair of adjacent columns, and based on the measure of similarity, a given percentage over other pairs of adjacent rows Identifying one of a pair of adjacent rows that differs by more than a pair of adjacent columns that differs by a given percentage more than other pairs of adjacent columns, and then the row of the identified pair of adjacent rows 9. The system of claim 8, wherein at least one direct label can be identified by identifying one of the identified pair of columns as including at least one direct label.   The document converter determines a characteristic vector for each cell in each pair of adjacent rows and each pair of adjacent columns, and the distance between cells in the same column in adjacent rows and between cells in the same row in adjacent columns. One of the distances can be determined based on the characteristic vector for each cell, and then the similarity measure can be determined by adding the distance across one of the adjacent row column and the adjacent column row. The system according to claim 9. A memory capable of storing an electronic document including table information comprising at least one table;
Convert at least one table into a linearized representation of at least one table based on the order of each table and also based on at least one label when at least one table includes at least one label A controller that can
With terminal.   The controller configures at least one table such that the at least one table has a predetermined order, localizes each label when the at least one table includes at least one label, and then at least one table Is based on at least one table configured to have a predetermined order, and at least one by linearizing based on each localized label when the at least one table includes at least one label 12. A terminal according to claim 11, wherein two tables can be converted.   Each table includes a plurality of cells organized in at least one row and at least one column, the controller constructs at least one table by identifying a natural order of the at least one table, and at least When one table has a natural order other than a predetermined order, at least one table having a natural order other than the predetermined order can be configured such that each at least one table has a predetermined order. The terminal according to claim 12.   The controller can determine a characteristic vector for each cell in the table, and determine a distance between adjacent cells based on the characteristic vector for each cell, and the controller can also determine the adjacent in the same row of the table. 14. The average distance between cell pairs and the average distance between adjacent cell pairs in the same column of the table can be determined, and then the natural order of the table can be identified based on these average distances. The terminal described in.   13. The terminal of claim 12, wherein the controller can configure at least one table such that the at least one table has one of a row-major order and a column-major order.   The controller can determine whether the terminal supports panning, and if the terminal supports panning, either convert at least one table or convert the electronic document without converting at least one table. The system of claim 11, wherein one of the choices to transfer can be received and the controller is capable of converting at least one table when the selection includes conversion of at least one table. A memory capable of storing an electronic document comprising table information comprising at least one table, each table comprising a plurality of cells organized in at least one row and at least one column;
A controller capable of identifying at least one of at least one span label and at least one direct label;
Each span label is associated with one of a plurality of columns and rows of the table, and each direct label is associated with one of a single column and single row of the table, and the controller includes: A label string for each cell of the table associated with at least one of the span label and the direct label is determined, whereby each span label and each direct label can be localized, and each label string includes the span label and the direct label. A terminal that contains at least one.   The terminal of claim 17, wherein the controller is capable of identifying at least one direct label based on a measure of similarity between one of adjacent rows and adjacent columns of the table.   The controller determines a measure of similarity between each pair of adjacent rows and one of each pair of adjacent columns, and based on this measure of similarity, a given percentage or more than other pairs of adjacent rows Identifying one of a pair of different adjacent rows and a pair of adjacent columns that differs by a given percentage more than other pairs of adjacent columns, and then the row of the identified pair of adjacent rows and adjacent columns The terminal of claim 18, wherein at least one direct label can be identified by identifying one of the identified pairs of columns as including at least one direct label.   The controller determines a characteristic vector for each cell in each pair of adjacent rows and each pair of adjacent columns, and the distance between cells in the same column in the adjacent row and the distance between cells in the same row in the adjacent column. Can be determined based on the characteristic vector for each cell, and then the similarity measure can be determined by adding the distance across one of the column of the adjacent row and the row of the adjacent column. Item 20. The terminal according to Item 19. In a method for converting table information for rendering to a terminal having a limited area display,
Receiving an electronic document comprising table information comprising at least one table, and at least one table based on the order of each table and also based on at least one label when at least one table contains at least one label Convert to a linearized representation of at least one table,
A method with a stage. The above stage of converting at least one table is:
Configuring at least one table such that the at least one table has a predetermined order;
Each label is localized when the at least one table includes at least one label, and the at least one table is based on at least one table configured to have the predetermined order, and the at least one table is Linearize based on each localized label when it contains at least one label,
The method of claim 21 comprising the steps of: Each table includes a plurality of cells organized in at least one row and at least one column, and the steps of constituting at least one table include:
Identifying the natural order of at least one table, and when the at least one table has a natural order other than a predetermined order,
Configuring at least one table having a natural order other than a predetermined order such that each at least one table has a predetermined order;
23. The method of claim 22, comprising the steps of: The above stage of identifying the natural order is for each table:
Determine the characteristic vector for each cell in the table,
Determine the distance between adjacent cells based on the characteristic vector for each cell;
Determining the average distance between adjacent cell pairs in the same row of the table and the average distance between adjacent cell pairs in the same column of the table, and identifying the natural order of the table based on these average distances;
24. The method of claim 23, comprising the steps of:   23. The method of claim 22, wherein configuring the at least one table comprises configuring the at least one table such that the at least one table has one of a row-major order and a column-major order. Decide if the terminal supports panning, and if the terminal supports panning,
Receiving a choice of either converting at least one table or transferring the electronic document without converting at least one table;
24. The method of claim 21, wherein the step of converting at least one table includes converting at least one table when the selection includes conversion of at least one table. A method for localizing at least one label of a table in an electronic document, the table comprising a plurality of cells organized in at least one row and at least one column, wherein:
Identifying at least one of at least one span label and at least one direct label, each span label being associated with one of a plurality of columns and rows of the table, and each direct label being a single column and single of the table Associated with one of the rows,
Determining a label string for each cell of the table associated with at least one of the span label and the direct label, thereby localizing each span label and each direct label, wherein each label string is at least one of the span label and the direct label; including,
A method with a stage.   28. The method of claim 27, wherein the step of identifying at least one direct label comprises identifying at least one direct label based on a measure of similarity between one of adjacent rows and adjacent columns of the table. The above steps for identifying at least one direct label include:
Determine a measure of similarity between each pair of adjacent rows and one of each pair of adjacent columns;
Based on the similarity measure, one of a pair of adjacent rows that differs by more than a given percentage from other pairs of adjacent rows and a pair of adjacent columns that differs by more than a given percentage than other pairs of adjacent columns And identifying one of the identified pair of rows of adjacent rows and the identified pair of columns of adjacent columns as including at least one direct label;
30. The method of claim 28, comprising the steps of: The above stage of determining a measure of similarity is:
Determine a characteristic vector for each cell in each pair of adjacent rows and each pair of adjacent columns;
One of the distance between cells in the same column in adjacent rows and the distance between cells in the same row in adjacent columns is determined based on the characteristic vector for each cell, and Add distance across one,
30. The method of claim 29, comprising the steps of: A computer program product for converting table information for rendering on a terminal having a limited area display, the computer program product comprising a computer readable storage medium storing a computer readable program code portion Wherein the computer readable program code portion is:
A first executable part for receiving an electronic document comprising table information comprising at least one table;
To convert at least one table to a linearized representation of at least one table based on the order of each table and also based on at least one label when the at least one table includes at least one label A second executable part of
A computer program product equipped with.   The second executable portion configures at least one table such that the at least one table has a predetermined order, localizes each label when the at least one table includes at least one label, and Thereafter, the at least one table is based on at least one table configured to have the predetermined order and is also linear based on each localized label when the at least one table includes at least one label. 32. A computer program product as claimed in claim 31.   Each table includes a plurality of cells organized in at least one row and at least one column, the second executable portion identifies a natural order of the at least one table, and the at least one table The at least one table having a natural order other than the predetermined order is configured such that each at least one table has a predetermined order when has a natural order other than the predetermined order. Computer program product.   The second executable portion determines a characteristic vector for each cell of the table, determines a distance between adjacent cells based on the characteristic vector for each cell, and further determines adjacent cell pairs in the same row of the table. Identify the natural order for each table by determining the average distance between and the average distance between pairs of adjacent cells in the same column of the table, and then identifying the natural order of the tables based on these average distances A computer program product according to claim 32.   33. The computer program product of claim 32, wherein the second executable portion comprises at least one table such that the at least one table has one of a row-major order and a column-major order. A third executable part that determines whether the terminal supports panning;
A fourth executable portion that receives a choice of either converting at least one table or transferring an electronic document without converting at least one table if the terminal supports panning;
32. The computer program product of claim 31, further comprising: converting the at least one table when the selection includes conversion of at least one table. A computer program product for localizing at least one label of a table in an electronic document, the table comprising a plurality of cells organized in at least one row and at least one column, and further comprising a computer readable program In a computer program product comprising a computer readable storage medium having a code portion stored thereon, the computer readable program code portion comprises:
A first executable portion for identifying at least one of at least one span label and at least one direct label, each span label being associated with one of a plurality of columns and rows of the table, and each direct The label is associated with one of the single column and single row of the table, and determines a label string for each cell of the table associated with at least one of the span label and the direct label, whereby each span label and A computer program product comprising a second executable portion for localizing each direct label, wherein each label string includes at least one of a span label and a direct label.   38. The computer program product of claim 37, wherein the first executable portion identifies at least one direct label based on a measure of similarity between one of adjacent rows and adjacent columns of the table.   The first executable portion determines a measure of similarity between each pair of adjacent rows and one of each pair of adjacent columns, and based on the similarity measure, Identify one of a pair of adjacent rows that differs by a given percentage and a pair of adjacent columns that differ by a given percentage more than other pairs of adjacent columns, and then the above identified pair of adjacent rows 40. The computer program product of claim 38, wherein one of the row and the identified pair of columns of adjacent columns is identified as including at least one direct label.   The first executable part determines a characteristic vector for each cell in each pair of adjacent rows and each pair of adjacent columns, and the distance between cells in the same column in adjacent rows and the same in adjacent columns One of the distances between the cells in a row is determined based on the characteristic vector for each cell, and then the similarity measure is calculated by adding the distance across one of the adjacent row column and the adjacent column row. 40. The computer program product of claim 39 to be determined.

Images