WO2014049057A1 - Procédé et système de fourniture séquentielle de données représentant des informations personnalisées, en particulier sous la forme de vidéos ou similaires, en particulier pour un programme de télévision personnalisé - Google Patents

Procédé et système de fourniture séquentielle de données représentant des informations personnalisées, en particulier sous la forme de vidéos ou similaires, en particulier pour un programme de télévision personnalisé Download PDF

Info

Publication number
WO2014049057A1
WO2014049057A1 PCT/EP2013/070079 EP2013070079W WO2014049057A1 WO 2014049057 A1 WO2014049057 A1 WO 2014049057A1 EP 2013070079 W EP2013070079 W EP 2013070079W WO 2014049057 A1 WO2014049057 A1 WO 2014049057A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
data
representing
videos
user
Prior art date
Application number
PCT/EP2013/070079
Other languages
German (de)
English (en)
Inventor
Jörg Wurzer
Original Assignee
Iqser Ip Ag
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Iqser Ip Ag filed Critical Iqser Ip Ag
Publication of WO2014049057A1 publication Critical patent/WO2014049057A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04H—BROADCAST COMMUNICATION
    • H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
    • H04H60/35—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users
    • H04H60/46—Arrangements for identifying or recognising characteristics with a direct linkage to broadcast information or to broadcast space-time, e.g. for identifying broadcast stations or for identifying users for recognising users' preferences
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
    • H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
    • H04N21/44213—Monitoring of end-user related data
    • H04N21/44222—Analytics of user selections, e.g. selection of programs or purchase activity
    • H04N21/44224—Monitoring of user activity on external systems, e.g. Internet browsing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q90/00—Systems or methods specially adapted for administrative, commercial, financial, managerial or supervisory purposes, not involving significant data processing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04H—BROADCAST COMMUNICATION
    • H04H20/00—Arrangements for broadcast or for distribution combined with broadcast
    • H04H20/26—Arrangements for switching distribution systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04H—BROADCAST COMMUNICATION
    • H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
    • H04H60/61—Arrangements for services using the result of monitoring, identification or recognition covered by groups H04H60/29-H04H60/54
    • H04H60/65—Arrangements for services using the result of monitoring, identification or recognition covered by groups H04H60/29-H04H60/54 for using the result on users' side
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04H—BROADCAST COMMUNICATION
    • H04H60/00—Arrangements for broadcast applications with a direct linking to broadcast information or broadcast space-time; Broadcast-related systems
    • H04H60/68—Systems specially adapted for using specific information, e.g. geographical or meteorological information
    • H04H60/73—Systems specially adapted for using specific information, e.g. geographical or meteorological information using meta-information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H04L67/564—Enhancement of application control based on intercepted application data
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/4508—Management of client data or end-user data
    • H04N21/4532—Management of client data or end-user data involving end-user characteristics, e.g. viewer profile, preferences

Definitions

  • the present invention relates to a method for the sequential provision of personalized information representing data, in particular in the form of videos or the like, in particular for a personalized television program.
  • the subject matter of the present invention is furthermore a data processing system for the sequential provision of personalized information representing data, in particular in the form of videos or the like, in particular for a personalized television program which is characterized by devices which are designed and / or arranged to carry out a method according to the invention.
  • the television known today is characterized by an editorially defined program.
  • the user only has the choice between many channels, which have different but fixed contents.
  • the Internet has created new conditions to break away from this editorially defined character.
  • an Internet TV could not prevail or it follows the model of conventional television. This is especially true of today's IP-TV.
  • Another attempt to break away from an editorially defined program is through online video rental. For payment of single content, a user will be provided.
  • the invention has for its object to provide a method for providing personalized information representing data, in particular in the form of videos or the like, in particular for a personalized television program, which assembles the content displayed personalized according to content and quality criteria. Furthermore, the content should be spontaneous contents of the user's wishes. The displayed content should therefore be dynamic and adaptable at any time.
  • the object is achieved by a method for the sequential provision of data representing personalized information, in particular in the form of videos or the like, which are not sequentially networked with one another, in particular for a personalized television program, which comprises the following method steps:
  • the invention is based on the knowledge that information-representing data (videos or the like) from one or more media libraries, for example video portals or the like, can be used to provide a spontaneous dynamic compilation of personalized contents to a user by suitable preprocessing can.
  • the data representing information is first analyzed content, the analysis for each of the information representing data is done individually.
  • the information representing information in parallel in an analysis chain, which consists of several sequential analysis steps, processed. Due to the preprocessing of the information representing data then a dynamic spontaneous user experience is possible.
  • a further realization of the invention is that a very simple and intuitive operation for navigation and interaction is required for a satisfactory user experience.
  • the output content may be spontaneously and dynamically adapted to a consumer's changing preferences to ensure this.
  • syntactic and semantic analysis for linking the information representing data (video or the like) and / or the information representing features (metadata);
  • information representing data (videos or the like) stored in libraries includes semantic metadata about title, author, subject, description, and / or Inform keywords (tags).
  • semantic metadata about title, author, subject, description, and / or Inform keywords (tags).
  • speech recognition a so-called speech-to-text method, the sound information of the data representing information can be extracted.
  • data representing information (videos or the like) in media libraries is usually already provided with a suitable classification as part of the metadata.
  • the evaluation of the quality of information representing data is advantageous.
  • a quality assessment and annotation of the information-representing data is therefore prescribed in a further embodiment of the invention.
  • the quality assessment and annotation of the information representing data is advantageously carried out optionally using the following parameters:
  • a production of a professional media agency is an indication of a higher quality and, consequently, a higher rating against information representing data created by amateurs;
  • Such methods or systems are disclosed, in particular, by WO 2005/050471 A2 and WO 2009/030247 A1.
  • the disclosures of WO 2005/050471 A2 and WO 2009/030247 A1 are hereby expressly cited as disclosure of the present application, in particular with regard to applications relating to a method for the sequential provision of personalized information representing data, in particular in the form of videos or the like, especially for a personalized television program.
  • the goal of such methods is the semantic networking of information representing data (video or the like).
  • information representing data which include an automatic moderation according to the content of the information representing data are advantageously short information representing data, in particular video sequences created, which are preferably used for a later moderation or overview of available content, in particular in the form of information representing data (videos or the like).
  • An advantageous embodiment of the invention is characterized in that the analysis and networking of data and / or features representing information (videos, metadata or the like) additionally comprises at least one of the following method steps:
  • CT (word) ( ⁇ TM ⁇ » ⁇ £ ⁇ log (word frequency (corpus) - word frequency (content object))
  • a correction of incorrect metadata which have arisen, for example, by spelling mistakes, can be made by a comparison with a vocabulary.
  • a supplement with a full text by an analysis, in particular a speech recognition (speech to text) of the sound information of the information representing data (Videos or the like) is advantageously carried out on the basis of a significance calculation.
  • the result (s) of such a significance calculation is advantageously used to automatically add further keywords to the information representing features (metadata) and to advantageously perform a pattern analysis with the most significant words.
  • the significance calculation of the sound information is advantageously carried out according to a computer linguistic method for a concept graph or category tree.
  • a formula which can be advantageously used for this purpose, wherein the parameters can be represented in particular by words in the sound information is:
  • CT word log (word frequency (corpus) - word frequency (content object))
  • a further advantageous embodiment provides alternatively for the application of the following formula, wherein the base value is freely selectable and preferably in a range of 1, 0 to 1, 0001:
  • the significance ⁇ is advantageously determined for each individual word contained in the sound information, in particular in order to determine the words with the greatest significance and to be able to supplement the information-representing features (metadata) on the basis of this result.
  • the content object indicates the object to be analyzed, in the present case, it is the sound information of the data representing information (video or the like).
  • the word frequency indicates the frequency of occurrence of a word within the content object.
  • the word count indicates the sum of all the different words contained in the content object.
  • the corpus is a compilation of information according to certain criteria, in this case it is a summation of all previously indexed content objects. Such methods or systems are disclosed in particular by WO 2009/13551 1 A1 and WO 2009/030247 A1.
  • WO 2009/13551 1 A1 and WO 2009/030247 A1 are hereby expressly cited as disclosure of the present application, in particular with regard to applications relating to the analysis and networking of data and / or features representing information (videos, metadata or the like ).
  • a further preferred embodiment of the invention is characterized in that the classification and annotation of information representing information (videos or the like) additionally comprises at least one of the following method steps:
  • the creation of a category tree via the information-representing features (metadata) is advantageously carried out via the calculation of the categories contained in the information-representing features (metadata) according to a method for calculating a category tree.
  • a method for calculating a category tree is disclosed in particular by WO 2009/13551 1 A1.
  • the disclosures of WO 2009/13551 1 A2 are hereby expressly cited as disclosure of the present applications, in particular with respect to applications concerning creation of a category tree.
  • the classification of the contents of a category tree is advantageously carried out by a comparison of the metadata with taxonomy or ontology, which are carried out using the literati of the hierarchized concepts.
  • An advantageous embodiment of the invention provides that the automatic production of data representing information (videos or the like) includes at least one of the following elements:
  • the use of a title and / or an author of data representing information (videos or the like) for an (animated) text or text layer is advantageously used.
  • the production is advantageously upgraded with the aid of an exemplary image, which can preferably be determined on the basis of face or panorama recognition.
  • the user advantageously indicates his intended results, for example for a TV magazine or a specific program, preferably by voice input, which translates a spoken text or words, so that they can be digitally processed further.
  • This automatically creates a query, also called a request below.
  • the selection or the call based on information defined by a user is advantageously carried out on the basis of the previous analysis and networking of the information representing data (videos or the like) and preferably results in a dynamic and spontaneous user experience for the user.
  • the query can be optionally interpreted in a further embodiment of the invention in order to be able to serve the user's intended results in a better and more targeted manner.
  • a method for automatic query interpretation can be used.
  • a method for at least partially automatic generation of a request to a data processing system with a database, wherein the data is based on a data model is used, which is used to select data from the database with a query language of the data processing system, a natural language query of a user, who by means of a user-input of concepts (user terms) acquired as a sequence of characters (strings), which terms semantically describe the user's intended results of the response of the data processing system to the request to the data processing system, the user's natural language query is analyzed and / or interpreted and translated into an inquiry in the query language, which is characterized in that the natural language query acquired as a string is decomposed into lexeme is set, these lexemes are processed sequentially according to their order in the string, with
  • a lexeme in the sense of the present invention is a unit of meaning, linguistic units, which can manifest itself in different word forms and thereby disregards the concrete form of the linguistic unity as well as the concrete syntactic function of the linguistic unity.
  • lexemes largely correspond to syntactic words, with the difference that, in particular, Asian languages and the like, which do not know any syntactically recognizable separation of words in a sentence, are also taken into consideration.
  • Syntactic words in the sense of the present invention are units which can be shifted within a sentence, exchanged by others and / or can be separated from one another by the insertion of further words.
  • Words in the sense of the present invention are, under syntactic aspect, the smallest displaceable and replaceable units of the sentence.
  • the use of the term lexeme in the context of the present invention takes into account that a given word can be reduced to a grammatical basic form and thus syntactic-grammatical variants can be compensated.
  • An inquiry or search request according to the invention is advantageously characterized in that the user or user intuitively describes the information he is looking for in an input field or text field with a plurality of words. This request is then broken down into its conceptual components and translated into an inquiry query language that goes beyond mere full-text search through a combination of keywords.
  • An advantageous embodiment of the invention provides that the user input is done by voice control or the like input help.
  • the data processing system advantageously has means for voice control and means for converting speech utterances into a sequence of characters.
  • An advantageous embodiment of the invention provides that it is checked for each currently to be used or processed lexeme, if there is a subsequent lexeme in the order, in which case it is checked whether it is the currently processed lexeme to is the name of a data source, a data class of the data model, or a data attribute of the data model.
  • this first checks whether the natural language request of the user is a plurality of terms. If this is the case, the first term is advantageously checked to see whether there is a data source, a data class, ie a semantic classification with regard to a content type, or a data attribute, ie metadata, with this term or term.
  • a further advantageous embodiment of the invention provides that in the case of the designation of a data source or a data class, the designation is transformed into a parameter of the query language for selecting data from the dataset, wherein the parameter is the selection of data on data with the designation of Data source or data class limited.
  • the data source is checked only for the first lexeme in order. A data source detected in this way then restricts the Search space for the request accordingly. For example, the term "Amazon" would only search in the data provided by the company Amazon, and if the first term corresponds to a data class, the search would be restricted to appropriately semantically classified content in terms of a content type "only search in the data classified as books by content type.
  • the information to the data processing system receives by using the available data model or in an optional embodiment of the invention advantageously of an ontology in which semantic content types representing data classes different names, so-called literate, contain or have in different languages.
  • the data processing system may further include or use dictionaries, vocabularies or the like as an ontology providing source of information.
  • a further advantageous embodiment of the invention provides that it is checked for each currently processed lexeme, if there is a subsequent lexeme in the order, in which case it is checked whether it is the name of a lexeme currently being processed Data source, a data class, or a data attribute of the data model.
  • a further advantageous embodiment of the invention provides that it is checked whether the lexeme currently to be processed or to be used is a noise word, that is to say a word with a low statement value, in which case the lexeme currently being processed is ignored and the following lexeme used or processed in the order.
  • the second term is thus analyzed or examined in the user's natural language query, where the second term may be a data class if the first term in the natural language query of the user has narrowed the query to a data source, or wherein the second term is a Search term if the first term in the user's natural language query corresponded to a data class representing a semantic content type, or where the second term may be a data attribute.
  • a further advantageous embodiment of the invention provides that in the case of the designation of a data attribute, the designation and the following lexeme is transformed in the order as a name-value pair into a parameter of the query language for selecting data from the dataset, wherein the parameter limits the selection of data to data with the name of the data attribute.
  • the information to the data processing system obtained by using the available data model or in an optional embodiment of the invention advantageously of an ontology in the data attributes (metadata) different names, so-called literate, contain or have in different languages.
  • the data processing system may further include or use dictionaries, vocabularies or the like as an ontology providing source of data attributes.
  • a term of the natural language request of the user is the designation of a data attribute
  • the query or search is limited to the data processing system by supplementing the query with the attribute value or the metadata. This restriction is then used for the following term in the user's natural language query.
  • the designation of a data class it is checked whether the designation is also the designation of a data attribute, in which case the lexeme currently to be processed and the subsequent lexeme in the order of name-value pair in a parameter of the query language for selecting data from the dataset is transformed, wherein the parameter for additional, alternative selection of data on data with the name of the data attribute confines.
  • Both parameters which describe the data class on the one hand and the data attribute on the other hand, are therefore linked as an OR link for querying the data.
  • a further advantageous embodiment of the invention is characterized in that in the case of the designation of no data source, no data class, no data attribute or no remaining lexeme in the order of the string the lexeme currently being processed into a parameter of the query language for selecting data from the database
  • the parameter limits the selection of data to data with the currently processed lexeme as a free text search.
  • the next term in the natural language query of the user may be a keyword that neither represents a data class nor a data attribute and is therefore advantageously used for a restrictive free text search or full text search.
  • the method according to the invention advantageously limits the complexity of the query in the query language to a single relation.
  • the user-side input is carried out using at least one input help, in particular for the semi-automatic generation of complex queries.
  • the user or user in the Wording of the request automatically guided.
  • the user-side input of the characters of the character string preferably the first input word
  • the user in the user-side input of the characters of the character string, preferably the first input word, to the input character string, the user becomes a list of possible data sources, data classes (semantic content types) and / or data attributes for a selection input made, for example, by clicking with a mouse pointer or the like displayed.
  • the user for the second word of the string of user-input is advantageously a list of possible data attributes (metadata) for example by clicking with a mouse pointer or the like selection input displayed.
  • the display of the possible selection inputs is advantageously carried out by visualization in the manner of a so-called “drop-down list", as it is known, for example, as functionality in so-called browsers.
  • the user-side input for complex or complex queries using a graphical user interface by creating a request tree.
  • GUI Graphical User Interface
  • the user or user can advantageously assemble a query tree or concatenate statements in the form of subject predicate object to be used in a list for the query.
  • a tree or graph representing such subject-predicate-object relationships is known, for example, from WO 2009/030288 A1, the disclosures of which are hereby expressly referenced.
  • the user advantageously first defines the root of the request tree.
  • the data class ie the semantic classification of a content type to select, preferably from a graphical user interface using a so-called mouse pointer or the like.
  • the user can add any number of attributes.
  • the user can also select from a list of possible data attributes that exist for the previously selected data class as a semantic content type in the data model.
  • the data attributes are optional by a logical "and" or For each attribute name or attribute name, advantageously, multiple attribute values may be indicated, which are linked by a logical "or". If the user does not select an attribute name, the stored keywords advantageously initiate a free text search or full text search for the selected data class.
  • the user can add restrictive links to the object thus defined, consisting of the data class forming the root of the request tree and / or the data attributes for this data class. For this, the user can also select these restricting links for the defined object from a list.
  • the user can limit the linkage himself by selecting a predicate, that is to say the reason for a link.
  • a predicate that is to say the reason for a link.
  • the data-processing system and / or the input assistant also provide him with a selection list again.
  • the user can save the request tree as a bookmark in a database or as a new data class of a semantic content type in an ontology, which ontology can then be used by the data processing system for further requests and / or request trees.
  • the completion of the interpretation of the query advantageously provides that a corresponding search index is based on the information representing data (videos or the like), which are located in the integrated libraries, is performed. As a result, a list of data representing information (videos or the like) is advantageously returned.
  • the method according to the invention is characterized in that the sorting and grouping of the results intended by the user is applied to at least one cluster and / or to the contents within at least one cluster.
  • clustering the results are grouped by content.
  • a clustering method advantageously builds up a content network or a graph which can be used accordingly.
  • Such methods or systems are disclosed in particular by WO 2009/030245 A1.
  • the disclosures of WO 2009/030245 A1 are hereby expressly cited as disclosure of the present application, in particular with respect to applications relating to a method of clustering.
  • Average data representing the representative information (video or the like).
  • the sorting of the clusters is preferably carried out on the basis of a heuristic, so that they are sorted in a specific order.
  • the content is sorted within a cluster according to at least one of the following criteria: so-called smart ranking based on a search optimization method;
  • the data representing information (video or the like) within a cluster is preferably sorted by a heuristic in an order.
  • Methods for search optimization are disclosed in particular by WO 2009/030248 A1.
  • the disclosures of WO 2009/030248 A1 are hereby expressly cited as disclosure of the present application, in particular with regard to applications relating to a so-called Smart Ranking based on a method for search optimization.
  • the sorting and grouping involves automatic production of information representing data (videos or the like) which includes automatic moderation according to the content of a cluster and / or the cluster.
  • the automatic moderation can consist of the data representing information (videos or the like), which comprise, for example, the highest quality rating, or a combination of already created automatic modalities.
  • An advantageous embodiment is characterized in that the input of the user for changing the output of data representing information is effected by a gesture and / or an actuation of a key.
  • the user can easily and intuitively navigate the output of information representing data (video or the like) on a terminal or a device designed to control it, for example a remote control.
  • the gesture causes the output to output information representing further information (video or the like) to the cluster prior to the execution of the gesture the operation of the key information representing active information (video or the like) is changed.
  • a further advantageous embodiment is characterized in that the gesture is carried out by swiping up or down and / or the operation of the key up or down, this being to the output of further information representing data (video or the like), which another cluster which is different from the cluster of data (videos or the like) representing prior to execution of the gesture or operation of the key of active information.
  • the present invention further proposes a data processing system for the sequential provision of personalized information representing data, in particular in the form of videos or the like, which are not sequentially interlinked, in particular for a personalized television program by devices which are designed and / or arranged to carry out a method according to the invention.
  • FIG. 1 shows a flow chart of an analysis and networking of information (video or the like) according to the invention in at least one dataset accessible via at least one data source;
  • FIG. 2 shows a flow chart of the sequence of a selection according to the invention based on information defined by a user and a sorting and grouping of the results intended by the user based on the analysis and networking of information according to the invention;
  • FIG. 3 shows a flowchart about the sequence of a further selection according to the invention based on information defined by a user, that is to say information based on a desired topic by a user, and a sorting and grouping according to the invention by the user intended results based on the analysis and networking of information;
  • FIG. 4 shows a schematic representation of an embodiment of a navigation concept according to the invention.
  • FIG. 1 shows the schematic sequence of an incremental analysis and networking of data representing information. It is carried out for each of the information representing data in at least one accessible via at least one data source database and therefore forms the basis for the provision of a particular personalized television program.
  • Step 1 starts the sequential performance of the analysis and networking for information representing information (videos or the like).
  • a video V, cf. Step 2 and it is checked in step 3 if the video V is new. If necessary, previously established relationships in networking can be checked for their validity. If the video is not new, it checks if the video V has already been seen, cf. Step 4. If this is the case, a semantic (usage) analysis is performed in step 6 to create or update relationships and / or weightings of the videos, for example, in the sense of machine learning. Thus, a trend tracking can be done. In step 7, this pass of the analysis and networking is terminated and a re-run for another data representing the information can be started. If video V was not yet viewed, cf. Step 4, the semantic (usage) analysis is omitted in step 6 and the process ends after query 4, (see end 5).
  • Step 8 is based on rules 9, for example, a correction of incorrect metadata, for example by typing errors by a Matching with, for example, a vocabulary.
  • metadata can be supplemented with synonyms of keywords by, for example, a comparison with thesaurus.
  • a supplement of a full text by a Speech recognition of the soundtrack of the video V done. The full text can be analyzed by a significance calculation performed on the words contained in the soundtrack. From this, additional keywords (tags) can be added automatically.
  • a classification and annotation of the video V 10 takes place.
  • the classification and annotation of the video V 10 can be carried out with the aid of ontology 1 1.
  • a calculation of the categories contained in the keywords (tags) can be carried out according to a method for calculating a category tree.
  • the classification can be carried out by comparing the keywords with ontology and / or taxonomy using the literates of hierarchical concepts.
  • the categories can be supplemented by a metadatum or attribute of the video V.
  • a quality evaluation and annotation of the video V 12 is performed on the basis of defined rules 13.
  • the evaluation of the quality of the videos is essential. Especially with publicly available sources there are strong quality fluctuations.
  • the quality assessment is based on various parameters such as the resolution of the video V, the image stability as an indication of a professional camera, the signal to noise ratio, the length of the video, the author or the source of the video and / or the quality of the metadata.
  • step 16 the process of sequentially performing the analysis and networking of information representing data (videos or the like) in step 7 ends.
  • 2 shows the basic sequence of the selection based on user-defined information and the sorting and grouping of the user-intended results based on the analysis and networking of information.
  • the sequence according to FIG. 2 is based on the results of the sequence according to FIG. 1, that is to say that the sequence according to FIG. 2 requires the results of the sequence according to FIG.
  • an input is initially made by a user entering a request A in step 19 which is to deliver the results intended by him.
  • a query Q is then created in step 20.
  • the content-related desire of the user may include, for example, a television magazine or specific program according to the current preferences of the user.
  • the input can preferably be done by voice input here.
  • step 21 the execution of the search with the query Q is performed under access to an index repository 22.
  • an index repository 22 a list of video content based on a selection based on user-defined information is created.
  • a content network 24 is analyzed by means of a clustering method.
  • the content network is the result of the incremental analysis and networking of information representing information in at least one dataset accessible via at least one datasource.
  • the results are grouped by content.
  • the clusters C1 to CN are sorted according to a heuristic Hi in step 25.
  • the clusters are sorted in an order on the basis of the heuristic. This can be done using the following criteria:
  • Anmoderationen in step 26 is based on a video repository 27.
  • the entire result or the individual clusters each receive a moderation.
  • the initial moderation can contain the best video contents of a cluster or a combination, ie a summary of the pre-produced moderation of individual video content.
  • the video contents within a cluster are also sorted according to a heuristic H2 in an order. This can be done by means of a smart ranking in the sense of a method for search optimization or based on the quality of the video content. This takes place in the method step provided with the reference numeral 28.
  • FIG. 3 shows a further embodiment of a selection according to the invention based on information defined by a user and sorting and grouping of the user-intended results based on the analysis and networking of information, wherein the creation of anmoderations in step 26 using video repositories 27 the sorting of the results in Ci according to a heuristic H 2 , cf. Step 28, done.
  • the remaining process steps are identical to those described in FIG.
  • the sequence according to FIG. 3 is based on the results of the sequence according to FIG. 1, that is to say that the sequence according to FIG. 3 requires the results of the sequence according to FIG.
  • FIG. 4 shows a particularly simple and innovative control or navigation concept which is suitable for use within the method according to the invention.
  • a user can easily and intuitively perform the control or navigation within the method according to the invention on a video terminal or a remote control 31.
  • the terminal 31 may for this purpose have a display device 32, in particular a touch-sensitive display device, as it is already widespread today.
  • the control or navigation takes place along two axes, one horizontal and one vertical.
  • control or navigation can also be performed by the operation on a terminal having keys.
  • control or navigation is based on the input of gestures.
  • a so-called swipe gesture to the left 33 or to the right 34 leads to alternative videos on the same thematic level of the current item, ie to videos within the cluster of the current reproduced output.
  • a swipe gesture up 35 or down 36 leads to a change to the previous or following subject area starting from the current reproduced output, ie to the previous or subsequent cluster of videos.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Social Psychology (AREA)
  • Databases & Information Systems (AREA)
  • Multimedia (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Business, Economics & Management (AREA)
  • Economics (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

La présente invention concerne un procédé de fourniture séquentielle de données représentant des informations personnalisées, en particulier sous la forme de vidéos ou similaires, en particulier pour un programme de télévision personnalisé, comprenant les étapes suivantes : 1. Analyse et chaînage de données représentant des informations (vidéos ou similaires) dans au moins un stock de données accessible par le biais d'au moins une source de données; 2. Sélection sur la base d'informations définies par un utilisateur; 3. Tri et groupage des résultats voulus par l'utilisateur sur la base de l'analyse et du chaînage des informations; 4. Sortie de données représentant les informations (vidéos ou similaires) sur la base du tri et du groupage; 5. Modification de la sortie des données représentant les informations (vidéos ou similaires) sur la base d'au moins une saisie de l'utilisateur concernant des résultats voulus modifiés, de préférence concernant une navigation dans les résultats voulus par l'utilisateur. L'invention concerne en outre un système de traitement de données en vue de la fourniture séquentielle de données représentant des informations personnalisées, en particulier sous la forme de vidéos ou similaires, en particulier pour un programme de télévision personnalisé, caractérisé par des dispositifs qui sont configurés et/ou adaptés pour mettre en œuvre un procédé selon l'invention.
PCT/EP2013/070079 2012-09-26 2013-09-26 Procédé et système de fourniture séquentielle de données représentant des informations personnalisées, en particulier sous la forme de vidéos ou similaires, en particulier pour un programme de télévision personnalisé WO2014049057A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
DE102012109096.3A DE102012109096A1 (de) 2012-09-26 2012-09-26 Verfahren zur sequenziellen Bereitstellung von personalisierte Informationen repräsentierenden Daten, insbesondere in Form von Videos und dergleichen, insbesondere für ein personalisiertes Fernsehprogramm
DE102012109096.3 2012-09-26

Publications (1)

Publication Number Publication Date
WO2014049057A1 true WO2014049057A1 (fr) 2014-04-03

Family

ID=49304911

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2013/070079 WO2014049057A1 (fr) 2012-09-26 2013-09-26 Procédé et système de fourniture séquentielle de données représentant des informations personnalisées, en particulier sous la forme de vidéos ou similaires, en particulier pour un programme de télévision personnalisé

Country Status (2)

Country Link
DE (1) DE102012109096A1 (fr)
WO (1) WO2014049057A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105704569A (zh) * 2016-03-08 2016-06-22 京东方科技集团股份有限公司 一种声音调节系统及声音调节方法

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102018212961A1 (de) 2018-08-02 2020-02-06 Robert Bosch Gmbh Verfahren und Vorrichtung zur Analyse eines Bildes

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005050471A2 (fr) 2003-11-22 2005-06-02 Wurzer Joerg Systeme et dispositif pour traiter des donnees
WO2009030248A1 (fr) 2007-09-03 2009-03-12 Iqser Ip Ag Détection de corrélations entre des données représentant des informations
WO2009030245A1 (fr) 2007-09-03 2009-03-12 Iqser Ip Ag Détection de corrélations entre des données qui représentent des informations
WO2009030247A1 (fr) 2007-09-03 2009-03-12 Iqser Ip Ag Détection de corrélations entre des données représentant des informations
WO2009030288A1 (fr) 2007-09-03 2009-03-12 Iqser Ip Ag Détection de corrélations entre des données représentant des informations
WO2009135511A1 (fr) 2008-05-08 2009-11-12 Iqser Ip Ag Création d’une arborescence des catégories sur le contenu d’un ensemble de données

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB0013702D0 (en) * 2000-06-07 2000-07-26 Meakes David Visual data interface

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005050471A2 (fr) 2003-11-22 2005-06-02 Wurzer Joerg Systeme et dispositif pour traiter des donnees
WO2009030248A1 (fr) 2007-09-03 2009-03-12 Iqser Ip Ag Détection de corrélations entre des données représentant des informations
WO2009030245A1 (fr) 2007-09-03 2009-03-12 Iqser Ip Ag Détection de corrélations entre des données qui représentent des informations
WO2009030247A1 (fr) 2007-09-03 2009-03-12 Iqser Ip Ag Détection de corrélations entre des données représentant des informations
WO2009030288A1 (fr) 2007-09-03 2009-03-12 Iqser Ip Ag Détection de corrélations entre des données représentant des informations
WO2009135511A1 (fr) 2008-05-08 2009-11-12 Iqser Ip Ag Création d’une arborescence des catégories sur le contenu d’un ensemble de données

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
EPO: "Mitteilung des Europäischen Patentamts vom 1. Oktober 2007 über Geschäftsmethoden = Notice from the European Patent Office dated 1 October 2007 concerning business methods = Communiqué de l'Office européen des brevets,en date du 1er octobre 2007, concernant les méthodes dans le domaine des activités", JOURNAL OFFICIEL DE L'OFFICE EUROPEEN DES BREVETS.OFFICIAL JOURNAL OF THE EUROPEAN PATENT OFFICE.AMTSBLATTT DES EUROPAEISCHEN PATENTAMTS, OEB, MUNCHEN, DE, vol. 30, no. 11, 1 November 2007 (2007-11-01), pages 592 - 593, XP007905525, ISSN: 0170-9291 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105704569A (zh) * 2016-03-08 2016-06-22 京东方科技集团股份有限公司 一种声音调节系统及声音调节方法
US10015433B2 (en) 2016-03-08 2018-07-03 Boe Technology Group Co., Ltd. Sound adjustment system and sound adjustment method

Also Published As

Publication number Publication date
DE102012109096A1 (de) 2014-03-27

Similar Documents

Publication Publication Date Title
Guo et al. Proactive wrangling: Mixed-initiative end-user programming of data transformation scripts
DE60208604T2 (de) Automatisches Verfahren zur Erzeugung von Image-buttons
DE69829074T2 (de) Identifizierung der sprache und des zeichensatzes aus text-repräsentierenden daten
DE60209572T2 (de) Verfahren und vorrichtung zur automatischen erkennung von datentypen für die datentypenabhängige verarbeitung
DE69432575T2 (de) Dokumentenerkennungssystem mit verbesserter Wirksamkeit der Dokumentenerkennung
DE112007000053T5 (de) System und Verfahren zur intelligenten Informationsgewinnung und -verarbeitung
DE102018007060A1 (de) Hervorheben von Schlüsselabschnitten eines Texts innerhalb eines Dokuments
DE102019000294A1 (de) Erstellen unternehmensspezifischer Wissensgraphen
DE102013003055A1 (de) Verfahren und Vorrichtung zum Durchführen von Suchen in natürlicher Sprache
US20050021545A1 (en) Very-large-scale automatic categorizer for Web content
DE102019001267A1 (de) Dialogartiges System zur Beantwortung von Anfragen
DE102013205737A1 (de) System und Verfahren zum automatischen Erkennen und interaktiven Anzeigen von Informationen über Entitäten, Aktivitäten und Ereignisse aus multimodalen natürlichen Sprachquellen
DE102017111438A1 (de) Api-lernen
EP2188742A1 (fr) Détection de corrélations entre des données qui représentent des informations
DE60101668T2 (de) Verfahren und gerät zum erzeugen eines auf einer formatvorlage basierten index für ein strukturiertes dokument
DE10308550A1 (de) System und Verfahren zur automatischen Daten-Prüfung und -Korrektur
DE112018005272T5 (de) Suchen von mehrsprachigen dokumenten auf grundlage einer extraktion der dokumentenstruktur
DE102018006241A1 (de) Gestalten eines Dokumentes unter Verwendung von intelligenten Merkmalsvorschlägen auf Grundlage einer textbasierten Analyse
DE112020005268T5 (de) Automatisches erzeugen von schema-annotationsdateien zum umwandeln von abfragen in natürlicher sprache in eine strukturierte abfragesprache
DE102021209171A1 (de) System zum wählen eines lernenden modells
Sperrle et al. Learning Contextualized User Preferences for Co‐Adaptive Guidance in Mixed‐Initiative Topic Model Refinement
WO2014049057A1 (fr) Procédé et système de fourniture séquentielle de données représentant des informations personnalisées, en particulier sous la forme de vidéos ou similaires, en particulier pour un programme de télévision personnalisé
DE212021000356U1 (de) Erzeugen einer gestuften Textformatierung für elektronische Dokumente und Anzeigen
EP1685505B1 (fr) Systeme de traitement de donnees
EP2856344A1 (fr) Génération de demandes à un système de traitement de données

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13773659

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13773659

Country of ref document: EP

Kind code of ref document: A1