WO2016175820A1

WO2016175820A1 - Interactive content selection

Info

Publication number: WO2016175820A1
Application number: PCT/US2015/028437
Authority: WO
Inventors: Lei Liu; Joshua Hailpern; Rares Vernica
Original assignee: Hewlett-Packard Development Company, L.P.
Priority date: 2015-04-30
Filing date: 2015-04-30
Publication date: 2016-11-03

Abstract

Examples disclosed herein relate to interactive content selection. In one implementation, a processor determines a concept associated with a piece of multimedia content. The processor may select from a set of multimedia content the piece of multimedia content to associate with a piece of educational content as interactive content based on a comparison of the determined concept to the educational content. The processor may output information related to a content creation including the selected piece of multimedia as interactive content associated with the educational content.

Description

Interactive Content Selection

BACKGROUND

[0001] Learning content may include print and digital resources. In some cases, learning content may be included within multimedia resources. For example, a student may learn with songs, videos, and computer games.

BRIEF DESCRIPTION OF THE DRAWINGS

[0002] The drawings describe example embodiments. The following detailed description references the drawings, wherein:

[0003] Figure 1 is block diagram illustrating one example of a computing system to select interactive content.

[0004] Figure 2 is a flow chart illustrating one example of a method to select interactive content.

[0005] Figure 3A is a diagram illustrating one example of selecting interactive content.

[0006] Figure 3B is a diagram illustrating one example of selecting segmented interactive content.

[0007] Figure 3C is a diagram illustrating one example of associating a question with selected interactive content.

DETAILED DESCRIPTION

[0008] In implementation, a processor automatically selects multimedia content to associate with educational content for user interaction. For example, the processor may select multimedia content to associate with educational content based on a comparison of the multimedia content to the educational content and create content by embedding information related to the selected multimedia content in the educational content such that a user interacts with the multimedia content. The processor may output information related to the created content, which may be, for example, an electronic book with links to additional multimedia content. The multimedia content may be automatically selected based on a concept associated with the educational content and the selected multimedia content. The multimedia content may be content that is retrieved based on a user interaction, such as by clicking a link to play a video or play a game. The multimedia content, such as a link, video, or game, may be selected based on a criteria related to its ability to augment learning associated with the educational content. In one implementation, a processor causes a user interface to be displayed that allows a user to provide criteria for the multimedia content selection and output. For example, an educator may select types of content or select a piece of content from a list of recommended multimedia content. Automatically selecting multimedia content for user interaction may greatly increase learning in a customized and scalable manner.

[0009] Figure 1 is block diagram illustrating one example of a computing system to select interactive content. For example, the computing system 100 may select multimedia content to associate with educational content in an interactive manner such as based on a similarity between concepts between the interactive content and the educational content. Additional criteria may also be applied, such as a criteria related to the degree to which the selected content augments the educational content or a difficulty level associated with the selected content. In one implementation, the processor takes into account output characteristics related to the selected content, such as resolution, size, or length. The computing system 100 includes a processor 101 , a machine-readable storage medium 102, and a storage 107.

[0010] The storage 107 may be any suitable storage for storing information accessible by the processor 101. The storage 107 may communicate directly or via network with the processor 101. The storage 107 includes multimedia content information 108. The multimedia content information 108 may include information related to multimedia content that may be associated with the educational content. The multimedia content may be, for example, a webpage, video, game, song, or image. In one implementation, a piece of multimedia content includes multiple types of content, such as a link including a video and audio options. The multimedia content information 108 may include information about the location of the content, such as a URL. The multimedia content information 108 may include information about a concept associated with the pieces of multimedia content. For example, a concept associated with a piece of multimedia content may be manually entered and/or automatically determined. The multimedia content information 108 may include additional information related to the output of the content, such as the size, resolution, or length.

[0011 ] The processor 101 may be a central processing unit (CPU), a semiconductor- based microprocessor, or any other device suitable for retrieval and execution of instructions. As an alternative or in addition to fetching, decoding, and executing instructions, the processor 101 may include one or more integrated circuits (ICs) or other electronic circuits that comprise a plurality of electronic components for performing the functionality described below. The functionality described below may be performed by multiple processors.

[0012] The processor 101 may communicate with the machine-readable storage medium 102. The machine-readable storage medium 102 may be any suitable machine readable medium, such as an electronic, magnetic, optical, or other physical storage device that stores executable instructions or other data (e.g., a hard disk drive, random access memory, flash memory, etc.). The machine-readable storage medium 102 may be, for example, a computer readable non-transitory medium. The machine-readable storage medium 102 may include multimedia content concept determination instructions 103, concept storage instructions 104, multimedia content selection instructions 105, and interactive content output instructions 106.

[0013] The multimedia content concept determination instructions 103 may include instructions to associate concepts with multimedia content. The multimedia content may be non-textual content, such as audio or video. In one implementation, the processor 101 creates text based on the multimedia content and determines a concept associated with the text. For example, a text may be created to represent the words in an audio file, and the processor may analyze the text to associate a concept with the text. The concept may be for example, a concept associated with a set of keywords and/or phrases. For example, the processor 101 or another processor may perform a machine learning method to determine a probability that a set of keywords is associated with a particular concept.

[0014] Multiple concepts may be associated with the same piece of multimedia content. In one implementation, the multimedia concept determination instructions 103 include instructions to segment a piece of multimedia content and associate a concept with each of the segments, such as based on a time period in a song or based on an audio link in a website having a different concept than a video link.

[0015] The concept storage instructions 104 include instructions to store the concept information in the storage 107 such that it is associated with the respective piece of multimedia content. For example, the multimedia content information 108 may include information about the multimedia content, such as a location, and information about a concept associated with the multimedia content.

[0016] The multimedia content selection instructions 105 includes instructions to select a piece of multimedia content based on the stored concept information compared to a piece of educational content. For example, a similarity between a piece of educational content and the multimedia content concepts may be determined. In one implementation, multimedia content is filtered based on additional criteria and then selected based on the concept similarity score, such as filtering to include all songs and then selecting a song based on the concept similarity score.

[0017] The interactive content output instructions 106 includes instructions to output information related to the selected multimedia content. For example, the content output instructions 106 may include instructions to store, display, or transmit information related to the selected multimedia content such that it may be provided as an interactive option to a user. In one implementation, the processor 101 creates a piece of content including the educational content and the selected multimedia content such that the educational content forms a base content from which the multimedia content may be launched due to user interaction. For example, the created content may be a digital book with a link to a video. Multiple pieces of multimedia content may be associated with the educational content or may be associated with different portions of the educational content. For example, the processor may automatically divide the educational content into segments based on concepts and associate interactive multimedia content with each of the segments.

[0018] Figure 2 is a flow chart illustrating one example of a method to select interactive content. For example, a processor may select multimedia content to form interactive content associated with educational content based on a concept determined to be related to the multimedia content. For example, the processor may determine a concept associated with a video and the degree of similarity between the determined concept and a concept associated with the educational content. The selected content may be output such that it supplements the educational content. For example, the educational content may include the selected content embedded and/or may include a link to the selected multimedia content such that a user of the educational content may interact with the multimedia content. The method may be implemented, for example, by the computing system 100 in Figure 1.

[0019] Beginning at 200, a processor determines a concept associated with a piece of multimedia content. The processor may determine a concept associated with the multimedia content in any suitable manner. The concept may be measured by a semantic topic, a taxonomy, or other content measurement, such as td-idf, that may be used to compare content. A method for concept determination may be selected based on the type of content, such as where a video segmentation method is used to determine a concept associated with video content. For example, edge information based video segmentation, image segmentation based video segmentation, and/or change detection based video segmentation may be used to segment a video and associate a concept with each segment. As another example, a concept may be determined for image content based on object identification, face recognition, image style, and object region segmentation.

[0020] In one implementation, a text method is used to determine a concept associated with non-textual content. For example, the processor may convert the content or a portion of the content to text and associate a concept with the text. For example, the text may be text within an image, text associated with a caption or metadata, and/or text associated with audio in audio or video content. The processor may remove a subset of the text, such as stop words prior to the concept analysis. In one implementation, the processor may determine a concept based on probabilities that the interactive content is associated with a concept based on the text or other information associated with the interactive content. For example, the concept may be associated .5 with a first concept and .8 with a second concept. The determined concept may be determined based on the concept with the highest score, a set of concepts with the highest scores, a profile based on a set of concepts, and/or other methods.

[0021] In one implementation, a processor creates a topic model to determine words and phrases associated with a concept, such as a topic model based on the probabilistic occurrence of terms in a text related to a particular topic. For example, a Probabilistic Latent Indexing method or a Latent Dirichlet Allocation (LDA) method may be used. The processor may cause the LDA method to be implemented in the following manner to create the topic model:

[0023] where β1:k is the topics and each βk is a distribution over the vocabulary. The topic proportions of the d-th document are θd where θd,k is the topic proportion for the topic k in document d. The topic assignments for the d-th document are zd where Zd.n is the topic assignment for the n-th word in document d. The observed words for the document d are Wd where Wd.n is the nth word in document d. The processor may simplify the topic model. For example, the processor may detect topics in the content and extract a subset of words, such as five words, with highest probability score βk represent the topic. In one implementation, the processor also takes into account phrases, such as based on N-grams, NER, and sentence phrase resolution. For example, a concept may have a probability of individual words appearing in a document and a probability of particular phrases.

[0024] In one implementation, the processor causes a user interface to be displayed to show information related to the determined concept such that a user may provide feedback by accepting the output, changing the determined concept, or selecting a sub-concept from a list of potential sub-concepts associated with the concept.

[0025] In one implementation, the processor determines multiple concepts associated with the interactive content such that different portions are associated with different concepts. For example, the processor may determine a concept per chapter, song of an audio track, or video track. The processor may automatically segment the content based on concept. For example, a video may be divided into time segments such that each time segment is associated with different concepts such as where a portion or length of the content is associated with a particular concept.

[0026] Continuing to 201 , a processor selects from a set of multimedia content the piece of multimedia content to associate with a piece of educational content as interactive content based on a comparison of the determined concept to the educational content. The educational content may be any suitable content such as a slide deck, book, or an article. The educational content may be textual and/or multimedia content, such as and e-book or video lecture. The processor may determine and/or access a concept associated with the educational content. The concept may be determined in the same or different manner as the concept determination method for the multimedia content. The concept associated with the educational content may be, for example, a single concept or a concept profile indicating weights associated with different concepts represented by the educational content. The educational concepts may be associated with multiple concepts such that different portions, for example, chapters, are associated with different concepts.

[0027] In one implementation, the processor performs a filtering method to the set of multimedia content. For example, the processor may filter based on size, resolution, concept, source, length, and/or type. The processor may access the filtering criteria from a storage or receive information related to the filtering criteria from user input. For example, the processor may cause a user interface to be displayed that receives user input related to selection criteria.

[0028] The processor may select the multimedia content in any suitable manner. For example, the processor may select the multimedia content with the same concept or highest likelihood of having the same concept as the educational content. In one implementation, the processor selects based on similarity of concept and other factors related to the multimedia content concept determined based historical learning data, such as data related to weighting the importance of concepts or which concepts are more frequently misunderstood. For example, frequently misunderstood concepts may be automatically determined based on analysis of aggregate student performance and/or analysis of the particular student's performance on previous courses, quizzes, or other assessments. Additional factors may be used to compare the multimedia content in addition to content, such as difficulty, understandability, and subject. For example, multimedia content with the same concept may be selected based on the additional factors. Other relevance information, such as same author, similar title, used together by previous users may also be taken into account.

[0029] The processor may generate a user interface to allow for user feedback on the selection. For example, the processor may select multimedia content with similarity scores with the educational content above a threshold, and a user, such as an educator, may select from the presented multimedia content which content is to be associated with the educational content.

[0030] Continuing to 201 , a processor outputs information related to a content creation including the selected piece of multimedia as interactive content associated with the educational content. For example, the processor may store, transmit, or display information related to the selection. The processor may output a content creation including the selected piece of multimedia content and the educational content. In one implementation, the processor creates multiple versions of the content creation such that a digital and print publication appear similar. For example, the print version may appear with the same text formatting with an icon or other item showing that interactive content is available on the digital version. In one implementation, an icon for interactive content may appear in the digital version but not the print version, and the remaining content may have the same formatting.

[0031] In one implementation, the processor selects multiple pieces of multimedia content. For example, links to the top N audio may be selected or the highest similarity of different types of content, such as the highest similarity video, highest similar audio content, and highest similarity image content. For example, links to the different types of content may be provided on the created content.

[0032] In one implementation, the multimedia content selection may be updated based on user provided information or based on information about user navigation. For example, content may be automatically reorganized based on user interest, and the interactive content may be updated accordingly. In one implementation, multiple pieces of interactive content are selected, and the processor provides a user an option to select one of the multiple pieces.

[0033] In one implementation, the processor associates a question with the selected piece of interactive content. The question may be related to the determined concept or more specifically to the selected multimedia content. The question may be an open ended question or a question with a preferred answer. The question may be determined in any suitable manner, such as based on a selection from a storage or based on question creation related to the specific multimedia content. The questions may be related to the multimedia content as a whole or a specific portion of the multimedia content. The processor may cause a user interface to be displayed to allow for user input related to the question generation. For example, the user interface may allow an educator to select from a set of recommended questions.

[0034] Figure 3A is a diagram illustrating one example of selecting interactive content. The storage 300 includes information related to pieces of multimedia content. For example, video A is related to concept 1 and song D is related to concept 4. The educational content 301 may be for example, print or digital text. Created content 302 includes both the educational content 301 and a link to video A 303 embedded in the educational content. For example, the created content 302 may be an e-book including a link to a video.

[0035] Figure 3B is a diagram illustrating one example of selecting segmented interactive content. For example, video A concept timeline 305 shows a timeline of the video A segmented by concept such that concept 6 is associated with the first minute and 20 seconds, concept 5 is associated with the time period between the first minute 20 second mark to the one minute 40 second mark, and concept 5 is associated with the time period between the one minute 40 second mark until the end of the video. The educational content 301 is related to concept 1 , and the link 307 included within the created content 306 includes a link to video A starting at the 1 minute 40 second mark.

[0036] Figure 3C is a diagram illustrating one example of associating a question with selected interactive content. For example, the created content 309 includes educational content 301 with a link 307 to a time segment of video A and includes interactive learning question 310. The interactive learning question may be, for example, related to concept 1 and/or related to the video portion included within the link. Automatically selecting multimedia content for interactive supplemental learning may improve learning outcomes in a scalable manner.

Claims

1. A computing system, comprising:

a storage to store information related to pieces of multimedia content; and a processor to:

determine concepts to associate with each of the pieces of interactive content, respectively;

store information related to the determined concepts in the storage;

select a piece of multimedia content to associate with a piece of educational content as interactive content based on a comparison of the educational content to the stored concept information associated with the piece of interactive content; and

output information related to a content creation including the educational content and interactive content including the selected piece of multimedia content.

2. The computing system of claim 1 , further comprising: outputting information related to a print publication such that the text of the print publication appears in the same formatting as a text portion of a corresponding digital publication.

3. The computing system of claim 1 , wherein the processor is further to segment a piece of multimedia content into multiple segments where each segment is associated with a different concept.

4. The computing system of claim 4, wherein the processor is further to associate a question with the selected piece of multimedia content.

5. A method, comprising:

determining, by a processor, a concept associated with a piece of multimedia content; selecting from a set of multimedia content the piece of multimedia content to associate with a piece of educational content as interactive content based on a comparison of the determined concept to the educational content; and

outputting information related to a content creation including the selected piece of multimedia as interactive content associated with the educational content.

6. The method of claim 5, further comprising segmenting the educational content into segments such that a different piece of multimedia content may be

associated with different segments.

7. The method of claim 5, further comprising segmenting the multimedia content based on divisions in concepts associated with the multimedia content.

8. The method of claim 5, further comprising filtering the set of multimedia content based on at least one of: size, resolution, concept, source, length, and type.

9. The method of claim 5, further comprising causing a user interface to be

displayed to allow a user to select criteria for the multimedia content selection.

10. The method of claim 5, further comprising determining a learning assessment question to associate with the selected multimedia content.

11.A machine-readable non-transitory storage medium comprising instructions

executable by a processor to:

select multimedia content to associate with educational content based on a comparison of concepts in the multimedia content to the educational content; create content by embedding information related to the selected

multimedia content in the educational content such that a user interacts with the multimedia content; and

output information related to the created content.

12. The machine-readable non-transitory storage medium of claim 11 , further comprising instructions to determine a concept associated with the multimedia content.

13. The machine-readable non-transitory storage medium of claim 11 , further

comprising instructions to determine at least one of a portion and length of the selected multimedia content to associate with the educational content.

14. The machine-readable non-transitory storage medium of claim 11 , wherein the processor is further to cause a user interface to be displayed to receive user input, wherein the user input is related to at least one of: criteria for the concept of the multimedia content and criteria related to the output characteristics of the selected multimedia content.

15. The machine-readable non-transitory storage medium of claim 11 , further

comprising instructions to update the multimedia content selection based on user feedback.