DESC:FIELD OF THE INVENTION
The invention relates to a system for interactive learning. The present invention further relates to a recommendation system for learning. Additionally the invention relates to a method of scoring/ranking the artefacts with relevance to the learner.

BACKGROUND OF THE INVENTION
The internet is abound with several learning artefacts including online courses, articles, blog posts, videos and podcasts around various topics. With this information overload, any knowledge worker is overwhelmed with choice. Collaborative filtering based inventions are in the prior art. These address the recommendation problem through the logic – “if person Y, who is part of X’s group / team is learning something, then X will also find it interesting.”. This tends to ignore personal preferences of X. There are 2 types of recommender systems – content based systems and collaborative filtering systems. This is an innovation in the former. Further, most of the prior art is around recommender systems for entertainment – for recommending a movie or album.

US20170193106A1 relates to a method and system for recommending content. The present teaching relates to recommending content by analyzing the streamed data. A request is received from a user requesting one or more recommendations from a set of items. A first distribution indicative of an interest distribution of the user in a plurality of topics is obtained. For each item, a second distribution indicative of a classification distribution of the item with respect to the plurality of topics is obtained. A score is estimated based on the first distribution and the second distribution, wherein the score indicates likelihood that the user is interested in the item. The scores associated with the set of items are ranked. The one or more recommendations are presented based on the ranked scores. In the reference invention, the recommendation is based on the overall topic interest profile built up for the user based on his previous activities. In the current invention, the recommendation system builds up an interest profile for the user with multiple topic clusters of interest. The recommendation offers a mix of learning artefacts aimed to satisfy nearly all the topic clusters of interest.

US20170154307A1 relates to a personalized data-driven skill recommendations and skill gap prediction. System and methods for generating personalized skill and learning recommendation are disclosed. A social networking system determines a list of skills not yet acquired by a first member. The social networking system determines a value associated with each skill. The social networking system selects one or more skills the list of skills based on the determined value of each skill. For each selected skill, the social networking system identifies at least one educational opportunity associated with learning the selected skill. The social networking system ranks the selected skills based on the determined value associate with each skill. The social networking system transmits a skill recommendation for an educational opportunity based on the selected skill rankings to a client system associated with the first member for display in a user interface. The reference invention identifies the skill gaps for a user based on his member profile and transmits skill recommendation. In the current invention, the recommendation is for content artefacts. If an artefact is tagged with a skill which the user has enrolled to, then its recommendation score is boosted.

US20140186817A1 relates to a ranking and recommendation of open education materials. A method of automatically ranking and recommending open education materials includes receiving a query. The method also includes calculating a content similarity measurement for each of multiple learning materials based on the query. The method also includes extracting multiple learning-specific features from the learning materials. The method also includes calculating one or more additional measurements for each of the learning materials based on the extracted learning-specific features. The one or more additional measurements are different than the content similarity measurement. The method also includes ranking each of the plurality of learning materials based on both the content similarity measurement and the one or more additional measurements. In the reference invention, the content similarity is based on other learning specific features. In the current invention, the content similarity is computed based on the overlap of topics.

US2009300547A1 relates to a recommender system for on-line articles and documents. A system and method for recommending on-line articles and documents to users is disclosed. The method provides an article widget user interface and a full-screen widget user interfaces to allow a user to rate articles, to preview articles, to filter articles based on category, article length, or other characteristics. A recommender system is configured to provide a continually refreshing list of recommended articles to the user via the user interfaces. The system comprises a module configured to monitor the user's explicit and implicit interactions with the user interfaces, and provides a refreshed list of recommended articles accordingly. The recommender system may be configured to use a package of approaches including rule-based, content-based or collaborative filtering approaches including Slope, Co-Visitation, Mwinnow and Clustering/Co-clustering. The reference invention recommends articles using a package of approaches including rule-based, content-based or collaborative filtering approaches. The current invention tracks user interaction with and provides recommendations from a variety of content types including articles, courses, blogs, videos and podcasts. In the current invention, the recommendation system builds up an interest profile for the user with multiple topic clusters of interest and simultaneously provides a mix of recommendations catering to the varied interests of the user.

Journal of Internet Technology Volume 20 (2019) No.6 relates to LCRec: Learning Content Recommendation (Wiki-based Skill Book). Knowledge skills in the ICT-industry always evolve. With the vast variety of jobs available, it is unlikely to educate students with skills to fit every job-requirement. This issue inspired us to develop the Learning Content Recommender (LCRec) for students to find appropriate learning contents based on required job-skills. In order to bridge the required skill for industry and academia, we have to work on IT job-skills and the Computer Science Curriculum 2013 (CS2013). Skills from 48 publicly available job searching websites are used to investigate what the industry needs. We carried out experiments among professionals, academics, and students to test the usefulness of LCRec, and evaluated the feedbacks. LCRec successfully used Knowledge Units from CS2013, Wikipedia, and essential skills from job hunting websites, to benefit entry-level job seekers for finding necessary learning contents to study. It is also convenient for academics to look at the skills needed in industries, and to consider enhancing the curriculum with new skills. The study result demonstrated that it is possible to bridge the gap (what learning contents are lacking) between the academia and the industry. The reference paper identifies skills associated with learning content by computing the similarity between the descriptions of the skill and the learning content. In the current invention, a skill is associated with multiple topics from the knowledge base. If learning content is related to any of the topics associated with a skill, it gets tagged with the skill. Additionally, the current invention uses the interlinks between the topics in the knowledge base. So even if the topics identified for a learning content are not directly related to any of the skills, if they have interlinks with a skill-related topic, they get tagged with the skill.

There exists a need for a recommendation system for learning. Further there exists a need for a method of scoring/ranking the artefacts with relevance to the learner. Further there exists a need for a tailored approach for the learning application.

OBJECTS OF INVENTION
Thus, according to the present invention, it is the primary object of the present invention to provide a system for interactive learning.

It is another object of the present invention to provide a method of scoring/ranking the artefacts with relevance to the learner.

It is another object of the present invention to provide a tailored approach for the learning application.

It is another object of the present invention, wherein the content repository system stores all the artefacts and also tags it by content type and skill.

SUMMARY OF THE INVENTION
One or more of the problems of the conventional prior art may be overcome by various embodiments of the present invention.

It is the primary aspect of the present invention to provide a recommendation system for interactive learning, comprising:
one or more processors;
a memory comprising one or more programs;
a data interface configured to receive a request from a user requesting one or more recommendations from a set of items/products/modules;
a content crawler unit configured to periodically crawl the web for new content and add to a content repository/database;
a content understanding unit configured to tag each artefact with relevant topics from a large, diverse knowledge base;
a skill ontology unit relating skills with relevant topics from the knowledge base;
a skill classifier unit configured to tag content artefacts with skills;
a skill proficiency unit;
a user profiling module configured to study the historical artefact consumption by a learner, cluster the topics of interest and thereby build a user interest profile;
a scoring module configured to estimate a score for a user-content artefact pair indicating the likelihood that the user is interested in the artefact;
a cluster analysis unit; and
a recommendation module configured to provide one or more recommendations based on the ranked scores,
wherein the content crawler unit periodically crawls the web for new content and adds it to the content repository and tags the content with relevant topics and skills, the recommendation module provides a personalized mix of learning modules/products based on interest profile of each learner/user selected on the basis of a recommendation score,
wherein the recommendation score is based on the overlap between the artefact topics and the topic cluster profile of the candidate/user content module/product and is based on learner/user relevance,
wherein the recommendation module tracks the preference of the learner/user for a particular content source and tailors the recommendations by boosting the learning modules/products from the preferred sources, and
the recommendation module tracks the learner/user preference for a particular type of content such as article, video, podcast and the like and tailors the recommendations.

It is another aspect of the present invention, wherein the content understanding unit tags each product with relevant topics from a diverse knowledge base using proprietary natural language processing methods that compares the content with the description of the topics from the knowledge base.

It is another aspect of the present invention, wherein the skill ontology unit is a proprietary resource relating all skills and competencies required from across industries with the relevant topics from the knowledge base with each other.

It is another aspect of the present invention, wherein the skill classifier unit tags / classifies each content product with relevant skills along with the proficiency level using the skill ontology unit and uses interlinks between the topics in the knowledge base.

It is another aspect of the present invention, wherein, the skill proficiency unit maps the actual skill level and desired skill level for each user/employee and comes up with a prioritized list of skills required for any user.

It is another aspect of the present invention, wherein the user profiling module identifies the topics of interest are identified for each user based on the historical consumption of content products, and groups the similar topics together and builds a user interest profile.

It is another aspect of the present invention, wherein the recommendation module represents a variety of content product types such as course, article, video, blog, podcast with topics from the knowledge base and the like.

It is another aspect of the present invention, wherein the recommendation is for content modules/products and if the content product is tagged with a skill in which the user has enrolled to, then the recommendation score is boosted.

It is another aspect of the present invention, wherein the recommendation module recommends a personalized mix of content modules/products based on the user’s topic interest profile.

It is another aspect of the present invention, wherein the recommendation module computes a similarity score which captures the overlap of topics between the content product and the user’s interest profile.

It is another aspect of the present invention, wherein the recommendation is for content products and if the content product’s type or source matches the user/learner’s preferred type or source, and then the recommendation score is boosted.

It is another aspect of the present invention, wherein the products/modules that are part of the prioritized skills are ranked higher, in the still rank.

It is another aspect of the present invention, wherein the contents in the content crawler unit comprises online courses, articles from known sources, videos from popular streaming channels, podcasts and the like.

It is another aspect of the present invention, wherein the content repository/database comprises content products of various content types such as courses, articles, blog posts, videos, podcasts from various sources and the like.

It is another aspect of the present invention, wherein the skill classifier unit uses interlinks between the topics in the knowledge base, in finding the relevant skills.

It is another aspect of the present invention, wherein the skill classifier unit is configured to identify skills even for content artefacts which are only peripherally related to the skill.

It is another aspect of the present invention to provide a method of working of the recommendation system for interactive learning, comprising the steps:
crawling of web source for new content products by the content crawler unit;
processing of newly acquired content products using the content understanding unit to tag them with topics;
tagging of topic tagged content products with relevant skills from the skill ontology unit using the skill classifier unit;
storing of skill and topic tagged content products in the content repository/database; and
recommendation system receiving request for recommendations for a user, comprising:
obtaining of user history and forming of the user interest profile by the user profiling module;
searching of the content repository and finding a list of content products best matched with the user profile by the recommendation module;
altering of the positions of content products in the recommendation list using additional parameters such as the user’s skill, content type and source preferences using the recommendation module; and
returning of final set of recommendations for the user.

BRIEF DESCRIPTION OF THE DRAWINGS
So that the manner in which the features, advantages and objects of the invention, as well as others which will become apparent, may be understood in more detail, more particular description of the invention briefly summarized above may be had by reference to the embodiment thereof which is illustrated in the appended drawings, which form a part of this specification. It is to be noted, however, that the drawings illustrate only a preferred embodiment of the invention and is therefore not to be considered limiting of the invention's scope as it may admit to other equally effective embodiments.
Figure 1: illustrates the block flow diagram of the content crawling, understanding and skill tagging mechanism according to the present invention.
Figure 2: illustrates a block flow diagram of the recommendation system according to the present invention.
Figure 3: illustrates the flow chart of the content crawling, understanding and skill tagging mechanism according to the present invention.
Figure 4: illustrates a flow chart of the working of the recommendation system according to the present invention.

DESCRIPTION FOR DRAWINGS WITH REFERENCE NUMERALS:
[100] Web/content source
[200] Content understanding unit
[300] Skill classifier unit
[400] Content repository/database
[500] Skill ontology unit
[600] User profiling module
[700] Recommendation module
[800] Recommendations
[10] Crawling of web source for new content products by the content crawler unit;
[20] Processing of newly acquired content products using the content understanding unit to tag them with topics
[30] Tagging of topic tagged content products with relevant skills from the skill ontology unit using the skill classifier unit;
[40] Storing of skill and topic tagged content products in the content repository/database
[50] Recommendation system receiving request for recommendations for a user
[60] Obtaining of user history and forming of the user interest profile by the user profiling module;
[70] Searching of the content repository and finding a list of content products best matched with the user profile by the recommendation module;
[80] Altering of the positions of content products in the recommendation list using additional parameters such as the user’s skill, content type and source preferences using the recommendation module
[90] Returning of final set of recommendations for the user

DETAILED DESCRIPTION OF THE INVENTION
It is to be understood that the present disclosure is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the drawings. The present disclosure is capable of other embodiments and of being practiced or of being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting.

The invention relates to a system for interactive learning. The present invention further relates to a recommendation system for learning. Additionally the invention relates to a method of scoring/ranking the artefacts with relevance to the learner.

Referring to Figure 1, the block flow diagram of the content crawling, understanding and skill tagging mechanism according to the present invention is illustrated. The content crawler obtains new content products from web sources. The content understanding unit processes these content products and tags them with topics from the knowledge base. The skill classifier unit then uses these topics along with the skill ontology to tag the content product with relevant skills. Finally, the content product tagged with topics and skills is stored in the content repository, ready for consumption by users.

Referring to Figure 2, the block flow diagram of the recommendation system according to the present invention is illustrated. Firstly, the user profiling unit accesses the content consumption history of the user to whom the recommendation is to be provided. A user interest profile is compiled and passed on to the recommendation module. The recommendation module accesses the content products from the content repository and identifies products best matched with the user interest profile. The recommendation module takes into account other aspects such as the user’s skill, content source and type preferences and forms a final set of recommendations to be provided to the user.

The recommendation system for interactive learning, comprises one or more processors, a memory comprising one or more programs, a data interface configured to receive a request from a user requesting one or more recommendations [800] from a set of items/products/modules, a content crawler unit configured to periodically crawl the web for new content and add to a content repository/database [400], a content understanding unit [200] configured to tag each artefact with relevant topics from a large, diverse knowledge base, a skill ontology unit [500] relating skills with relevant topics from the knowledge base, a skill classifier unit [300] configured to tag content artefacts with skills, a skill proficiency unit, a user profiling module [600] configured to study the historical artefact consumption by a learner, cluster the topics of interest and thereby build a user interest profile, a scoring module configured to estimate a score for a user-content artefact pair indicating the likelihood that the user is interested in the artefact, a cluster analysis unit and a recommendation module [700] configured to provide one or more recommendations [800] based on the ranked scores. The content crawler unit periodically crawls the web for new content and adds it to the content repository [400] and tags the content with relevant topics and skills, the recommendation module [700] provides a personalized mix of learning modules/products based on interest profile of each learner/user selected on the basis of a recommendation score. The recommendation score is based on the overlap between the artefact topics and the topic cluster profile of the candidate/user content module/product and is based on learner/user relevance. The recommendation module [700] tracks the preference of the learner/user for a particular content source [100] and tailors the recommendations [800] by boosting the learning modules/products from the preferred sources [100]. The recommendation module [700] tracks the learner/user preference for a particular type of content such as article, video, podcast and the like and tailors the recommendations [800].
The content understanding unit [200] tags each product with relevant topics from a diverse knowledge base using proprietary natural language processing methods that compares the content with the description of the topics from the knowledge base. The skill ontology unit [500] is a proprietary resource relating all skills and competencies required from across industries with the relevant topics from the knowledge base with each other. The skill classifier unit [300] tags / classifies each content product with relevant skills along with the proficiency level using the skill ontology unit [500] and uses interlinks between the topics in the knowledge base. The skill proficiency unit maps the actual skill level and desired skill level for each user/employee and comes up with a prioritized list of skills required for any user. The user profiling module identifies the topics of interest are identified for each user based on the historical consumption of content products, and groups the similar topics together and builds a user interest profile. The recommendation module [700] represents a variety of content product types such as course, article, video, blog, podcast with topics from the knowledge base and the like.

The recommendation [800] is for content modules/products and if the content product is tagged with a skill in which the user has enrolled to, then the recommendation score is boosted. The recommendation module [700] recommends a personalized mix of content modules/products based on the user’s topic interest profile. The recommendation module [700] computes a similarity score which captures the overlap of topics between the content product and the user’s interest profile. The recommendation [800] is for content products and if the content product’s type or source [100] matches the user/learner’s preferred type or source [100], and then the recommendation score is boosted. The products/modules that are part of the prioritized skills are ranked higher, in the still rank. The contents in the content crawler unit comprises online courses, articles from known sources [100], videos from popular streaming channels, podcasts and the like. The content repository/database [400] comprises content products of various content types such as courses, articles, blog posts, videos, podcasts from various sources [100] and the like.
The skill classifier [300] unit uses interlinks between the topics in the knowledge base, in finding the relevant skills. The skill classifier unit [300] is configured to identify skills even for content artefacts which are only peripherally related to the skill.

Referring to Figure 3, the flow chart of the content crawling, understanding and skill tagging mechanism according to the present invention is illustrated. Referring to Figure 4, a flow chart of the working of the recommendation system according to the present invention is illustrated. The method of working of the recommendation system for interactive learning, comprising the steps: crawling [10] of web source for new content products by the content crawler unit; processing [20] of newly acquired content products using the content understanding unit [200] to tag them with topics; tagging [30] of topic tagged content products with relevant skills from the skill ontology unit [500] using the skill classifier unit [300]; storing [40] of skill and topic tagged content products in the content repository/database [400]; and recommendation system receiving [50] request for recommendations for a user, comprising: obtaining [60] of user history and forming of the user interest profile by the user profiling module [600]; searching [70] of the content repository [400] and finding a list of content products best matched with the user profile by the recommendation module [700]; altering [80] of the positions of content products in the recommendation list using additional parameters such as the user’s skill, content type and source preferences using the recommendation module [700]; and returning [90] of final set of recommendations for the user.

Algorithm accounts for the semantic context of the words in the content while identifying the relevant topics. The algorithm used to identify the topics relevant to a content product, functions as follows: The topic descriptions in the knowledge base are represented both as a collection of words and as a distribution over latent topics. The latent topics used in the latter representation are obtained using a dimensionality reduction method. For a given content product with a textual description, the algorithm forms lists of the most related topics using each of the two representations. In each list, the topics are ordered by a score which captures the degree of relatedness. These two lists are then combined with appropriate weighting and filtering conditions. The content product is tagged with the top topics from the combined list.

The content similarity is computed based on the overlap of topics. The recommendation module recommends content modules/products based on the user’s topic interest profile. The recommendation module builds up an interest profile for the user with multiple topic clusters of interest and simultaneously provides a mix of recommendations catering to the varied interests of the user. The recommender module uses a proprietary mathematical method to rank each content for each user using the following parameters such as skill rank, content type rank and similarity score.

Each product database/artefact
Polysemy is the capacity for a word or phrase to have multiple meanings, usually related by contiguity of meaning within a semantic field.

The present invention uses different systems – a content crawler system, a content understanding system, a skill ontology, a skill classifier system, a content repository, a skill proficiency system and a recommender system. The content crawler system periodically crawls the web for new online courses, articles from known sources, videos from popular streaming channels and podcasts and adds it to the content repository. The content repository system not only stores all the artefacts but also tags it by content type and skill. The content understanding system tags each artefact with relevant topics from a large, diverse knowledge base. This is done using a proprietary NLP algorithm that compares the content with the description of the topics from the knowledge base. The algorithm accounts for the semantic context of the words in the content while identifying the relevant topics thereby avoiding the problem of polysemy.

The skill ontology is a proprietary resource relating all skills / competencies required from across industries with the relevant topics from the knowledge base and with each other. The skill classifier system tags / classifies each content artefact with relevant skills using the skill ontology. Additionally the skill classifier system uses interlinks between the topics in the knowledge base. So even if none of the topics tagged with a content artefact are related to any skill, if the topics have interlinks with a skill-related topic, the artefact can be tagged with the skill. Thereby the skill classifier is even able to identify skills for content artefacts which are only peripherally related to the skill. The skill proficiency system maps the actual skill level and desired skill level for each employee and thus comes up with a prioritized list of skills required for any user (the skills that have the largest gap are prioritized first)

Based on the historical consumption of content artefacts, topics of interest are identified for each user. A cluster analysis is performed to group together similar topics. Thus an interest profile is built up for the user with topic clusters of interest. The recommendation engine provides a personalized mix of learning artefacts aimed to satisfy the varied interest profile of each learner. A recommendation score is computed for each content artefact which is composed of the following elements:

? A similarity score based on the overlap between the topic clusters and the topic profile of the candidate content artefact.
? In addition to the learners’ topics of interest, the recommendation engine also tracks any preference the learner may have for a particular content source and tailors the recommendations by boosting learning artefacts from the preferred sources.
? Similarly, the recommendation engine also tracks learner preference for a particular type of content i.e. article, video or podcast and tailors the recommendations
? Additionally artefacts tagged with any of the prioritized skills are ranked higher

Advantages:
• More relevant learning recommendations than a random web search
• Tailored recommendations to a particular job role

Although, the invention has been described and illustrated with respect to the exemplary embodiments thereof, it should be understood by those skilled in the art that the foregoing and various other changes, omissions and additions may be made therein and thereto, without parting from the spirit and scope of the present invention. ,CLAIMS:WE CLAIM:
1. A recommendation system for interactive learning, comprising:
one or more processors;
a memory comprising one or more programs;
a data interface configured to receive a request from a user requesting one or more recommendations [800] from a set of items/products/modules;
a content crawler unit configured to periodically crawl the web for new content and add to a content repository/database [400];
a content understanding unit [200] configured to tag each artefact with relevant topics from a large, diverse knowledge base;
a skill ontology unit [500] relating skills with relevant topics from the knowledge base;
a skill classifier unit [300] configured to tag content artefacts with skills;
a skill proficiency unit;
a user profiling module [600] configured to study the historical artefact consumption by a learner, cluster the topics of interest and thereby build a user interest profile;
a scoring module configured to estimate a score for a user-content artefact pair indicating the likelihood that the user is interested in the artefact;
a cluster analysis unit; and
a recommendation module [700] configured to provide one or more recommendations [800] based on the ranked scores,
wherein the content crawler unit periodically crawls the web for new content and adds it to the content repository [400] and tags the content with relevant topics and skills, the recommendation module [700] provides a personalized mix of learning modules/products based on interest profile of each learner/user selected on the basis of a recommendation score,
wherein the recommendation score is based on the overlap between the artefact topics and the topic cluster profile of the candidate/user content module/product and is based on learner/user relevance,
wherein the recommendation module [700] tracks the preference of the learner/user for a particular content source [100] and tailors the recommendations [800] by boosting the learning modules/products from the preferred sources [100], and
the recommendation module [700] tracks the learner/user preference for a particular type of content such as article, video, podcast and the like and tailors the recommendations [800].

2. The recommendation system for interactive learning as claimed in claim 1, wherein the content understanding unit [200] tags each product with relevant topics from a diverse knowledge base using proprietary natural language processing methods that compares the content with the description of the topics from the knowledge base.

3. The recommendation system for interactive learning as claimed in claim 1, wherein the skill ontology unit [500] is a proprietary resource relating all skills and competencies required from across industries with the relevant topics from the knowledge base with each other.

4. The recommendation system for interactive learning as claimed in claim 1, wherein the skill classifier unit [300] tags / classifies each content product with relevant skills along with the proficiency level using the skill ontology unit [500] and uses interlinks between the topics in the knowledge base.

5. The recommendation system for interactive learning as claimed in claim 1, wherein, the skill proficiency unit maps the actual skill level and desired skill level for each user/employee and comes up with a prioritized list of skills required for any user.

6. The recommendation system for interactive learning as claimed in claim 1, wherein the user profiling module identifies the topics of interest are identified for each user based on the historical consumption of content products, and groups the similar topics together and builds a user interest profile.

7. The recommendation system for interactive learning as claimed in claim 1, wherein the recommendation module [700] represents a variety of content product types such as course, article, video, blog, podcast with topics from the knowledge base and the like.

8. The recommendation system for interactive learning as claimed in claim 1, wherein the recommendation [800] is for content modules/products and if the content product is tagged with a skill in which the user has enrolled to, then the recommendation score is boosted.

9. The recommendation system for interactive learning as claimed in claim 1, wherein the recommendation module [700] recommends a personalized mix of content modules/products based on the user’s topic interest profile.

10. The recommendation system for interactive learning as claimed in claim 1, wherein the recommendation module [700] computes a similarity score which captures the overlap of topics between the content product and the user’s interest profile.

11. The recommendation system for interactive learning as claimed in claim 1, wherein the recommendation [800] is for content products and if the content product’s type or source [100] matches the user/learner’s preferred type or source [100], and then the recommendation score is boosted.

12. The recommendation system for interactive learning as claimed in claim 1, wherein the products/modules that are part of the prioritized skills are ranked higher, in the still rank.

13. The recommendation system for interactive learning as claimed in claim 1, wherein the contents in the content crawler unit comprises online courses, articles from known sources [100], videos from popular streaming channels, podcasts and the like.

14. The recommendation system for interactive learning as claimed in claim 1, wherein the content repository/database [400] comprises content products of various content types such as courses, articles, blog posts, videos, podcasts from various sources [100] and the like.

15. The recommendation system for interactive learning as claimed in claim 1, wherein the skill classifier [300] unit uses interlinks between the topics in the knowledge base, in finding the relevant skills.

16. The recommendation system for interactive learning as claimed in claim 1, wherein the skill classifier unit [300] is configured to identify skills even for content artefacts which are only peripherally related to the skill.

17. A method of working of the recommendation system for interactive learning, comprising the steps:
crawling [10] of web source for new content products by the content crawler unit;
processing [20] of newly acquired content products using the content understanding unit [200] to tag them with topics;
tagging [30] of topic tagged content products with relevant skills from the skill ontology unit [500] using the skill classifier unit [300];
storing [40] of skill and topic tagged content products in the content repository/database [400]; and
recommendation system receiving [50] request for recommendations for a user, comprising:
obtaining [60] of user history and forming of the user interest profile by the user profiling module [600];
searching [70] of the content repository [400] and finding a list of content products best matched with the user profile by the recommendation module [700];
altering [80] of the positions of content products in the recommendation list using additional parameters such as the user’s skill, content type and source preferences using the recommendation module [700]; and
returning [90] of final set of recommendations for the user.