 This is the OER team, I am Mahesh Sharma, my friend Shankul and my friend Nikhil. So basically the project, Development of Open Educational Resources Repository. So first thing first, let's get an idea of what is an OER. So it is a free access, these are the freely accessible and openly licensed documents. There can be any media resources, which can be useful for learning, teaching, like assessing or sometime for research purposes. They consist of some copy-left licenses, like that's a basic knowledge. Second thing, so on what thing we built our repository, alright. So on what we built our repository, we built our repository on dSpace and as most of us are not so familiar with dSpace, so first thing like what is dSpace? So dSpace is an open source repository, it's a Java-based application basically that allows us to capture, store, index, preserve and distribute our digital material. Basically it provides us the interfaces for administration, deposit, ingest, search and to access the tools, access the materials we provide them. Second thing like I'll try to explain what all we did in this project through the architecture of dSpace. So the architecture like it has the three basic layers, application layer, business logic layer and storage layer. In the application layer like it is basically so that we can communicate with the world outside the individual dSpace installed in a PC. So in the application layer, we worked on the statistics tools like we integrated our dSpace with ELK stack, after that we changed the web UI also and in the business logic layer, we worked on the search and the browse tools under the discovery implementation. We changed the authorization methods also, we provided some administration toolkit like in the profile page and the input page and we changed some E-Person group manager like stuff in which we changed the profile page and the input page fields. In the storage layer, basically we just integrated the JDBC plugin to integrate our database with LogStash, the PostgreSQL database with LogStash. Now structure of the dSpace, so this is like basically how the files are stored in our repository. In dSpace we have communities, a system can have zero to n communities and each community can have sub communities, this can also go to n levels. So basically a community can have sub communities and any community or sub community can have collections. So those collections only have items, an item can only belong to a collection and this item is only our resource, it can be the video link or the video itself or the PDF. So now coming to the implementations we made, first I'll start with the updating profile page. So basically by default the dSpace provides us with the dSpace profile page, contains the fields like name, contact number and some other fields like an optional password update field also. But as to make it look like more educational repository, we made a few changes and added these additional fields which were like which we used in our Kibana visualization later on. So approach for that is like we know that each user accessing the OER, for an each user accessing the OER we have an E-person object created in the back end of the system. So basically what we did is we had to update the schema for that E-person object and the following changes were made for this. Like we changed some XML and JSP files for creating new fields at the front end for fetching the data entered by users respectively. Some Java files were changed so to update the E-person object content. Some queries were made in the, additional queries were made in PostgreSQL database files to allow the storage of the newly updated fields. Second change we made was in the input page. Again by default the input page has some fields, some fields but and it uses the Dublin core metadata schema. So for adding our additional fields we had to make changes in the Dublin core metadata schema and for the same like, for the same we had to do the changes like first we updated the configuration files, the input forms configuration files and to make it reflect in the d-space itself we had to go in d-space and like only this administrator can make the changes in the Dublin core metadata schema there. So and apart from this we introduced some IEEE form and metadata fields too. Now my friend Shankul will conduct. So I have been discussing about the discovery module over here. So we know that this d-space, Mahesh have told us that it is a framework. So framework has multi functionalities been implemented, predefined. So this discovery module is also been predefined in the d-space module. So what is this? This d-space module have functionality provided for browsing for any of the resources for providing any of the, for browsing for any of the course module or domain. So it is also for face searching. What is this face is searching? We have gone to Amazon or something, Mintra like that. So we have searched out some clothes. So what we do? We search for some color, we search for some size. So that is basically this first searching. What is this? We categorize our search into different fields means whatever subject we want to search, whatever the contributor we want to search. This is being filters have been combined to provide us the best result into some query. So d-space, this discovery module has been implemented from the solar, which has been implemented the Apache Lucene framework. So this Apache Lucene framework uses the inverted index functionalities. What is this inverted index functionality? This is like, we have seen the, the index of our book. So we, in the book, what do we do? We not search for that word in the whole of the book. Like, we search for that word in the index and that we directly go there. Because for the first implementation, it is very hard to search for a word. So that is very much hard. So this is being, inverted index functionality has been implemented in the solar, which is then used by the discovery module. So what, the discovery module, we have implemented three things. First is the tagloud. What is tagloud? We have seen it multiple times in our lives. What is this? This is a visual representation in which each of the key word or subject has been, has been provided some weight, weight in the form of font, weight in the form of what is the color of this is, means any of the thing which has been used more or which has been accessed more should be provided more weightage. So that it would be like more of been attracted to the user which has been using that thing. So that is the tagloud we have implemented. Now second thing is, we have added some filters, means as Mahesh have told us, we have changed some metadata registry for that. We have implemented some other IEEE registry also. So for that, we have changed some filters and faces in the discovery module first, so that we are well defined with the community as well as any of the collection of home pages. And as well as we have implemented the recent submission module. So we want that if any of the course has been added to the repository, it has been, it should be shown into the home page. So what is this recent submission is, means if any of the courses had been added into the whole domain of the D-space, it will be shown into the home page of the D-space. As well as for each of the community pages, community pages, I am talking about the domains, means computer science domain or electrical domain. So in the electrical or computer science domain also, if any of the courses had been added, it will be separately shown for that community. This is the recent submission which we have implemented in the D-space. Now talking about the different thing, which is the OER dashboard. So D-space doesn't provide us some functionality about the dashboard for each of the user. In the modern world, we want that each user should have that thing that whatever the course you want to loan or want to have, should be separate with each of the user. That's why we have implemented each user's dashboard. So what is the functionalities methodology? First is the authorization. Basically, we want to authorize basically that user has the authorization to access his dashboard. If not, we'll be redirecting it to the login page. Secondly is the access. Access means providing access to means delete or to see that course or to add some course. So this is basically the OER dashboard. So I have told you the implementation. So what basically in the implementation is there means for adding any of the courses, we'll be using Hibernate means we have created some Pojo classes. These Pojo classes are being like get a set of functions for the user to see whatever the course you want to add it to is a dashboard. And then that is mapped to the D-OER class. Then D-OER class will be using Hibernate to just map it down to the post-grace SQL which we have using. And some programming in the GSP server for the request objects we have been seeing. Now in the like we have also implemented the means there are various resources in the OER like the video or the PDF. We want it to be displayed in the D-Space or in the R-OER only. Why? We don't want it to download each time we see or redirect it to the YouTube because in the metadata which we are having from the server, we have YouTube links and we want it to be displayed into the D-Space only. So that's why we have implemented video for video we have used video.js and YouTube.js as content-driven network server. So we have implemented that in a different server and we'll be using in our D-Space module. And for the PDF we are using iFrames and in those iFrames we'll be showing the PDFs in various courses or domains. And the additional modification we have done means D-Space doesn't have support for the MP4 videos so it has support for the MIME field so we have changed that and also we have slightly modified some UI on the D-Space to suit our OER repository which we want to develop. Okay so now we'll be talking about the ELK stack this will be talking about the locks and Nikhil will continue it. Now next we have ELK stack so what is ELK stack? ELK stack is the acronym for three open source projects, Elasticsearch, Lockstash and Kibana. So now moving straight forward to Elasticsearch. Elasticsearch is an open source full-text search engine which is document based and it uses JSON format to represent the data. Elasticsearch uses indexing for fast fetching of data. Now lockstash, lockstash is like a data pipeline that takes the unstructured data as input process data converts that unstructured data into structured form and later feeds that process data to other applications like Elasticsearch. The major three components of lockstash are inputs, filters, outputs. Next we have is Kibana. Kibana is a visualization tool which is used to represent the Elasticsearch data in the form of visual formats. The sum of the basic features of Kibana are its user-friendly interface, multiple customizable dashboards and real-time searching of index data. So why do we need Elasticsearch? Together all the different components of Elasticsearch provide a simple yet efficient solution for lock management and analytics. Lock management and analytics include collection of our lock data files, cleaning of the lock data, converting the unstructured lock data file to structured form, then analyzing our process data and obtaining the results. As you can see in this diagram, what we are doing, we have taken our lock file as input, containing the unstructured data. That unstructured data is passed to lockstash data pipeline. What lockstash will do, it will convert that unstructured data to structured form and will pass that structured data to Elasticsearch for further indexing. Once the indexing is done on the structured data, it is used by Kibana for visualization task. Requirement of a project was to create the visualization representing the information such as date, time, user login ID, IP address of user, accessing our repository. It also included information about the courses, domain and subjects visited by the user and number of times they were visited. The main task of our project was to read the lock files for that we have used a grog filter provided by lockstash. What grog filter does is it reads the lock file line by line and matches the each line with the pattern we are looking for. Once the match is found, the data is extracted and stored in structured form. For defining grog pattern, we have used regular expression. Once after reading the lock files, we got the data containing course ID, domain ID and subject ID. But to make Kibana visualization more precise, we needed their names also that were stored in the postgres database. So for that we have used JTBC driver which established a connection between our lockstash and postgres database to fetch the further information. Once we got that information stored in the database, we used the translate filter to create a mapping between our course ID, course name, domain ID, domain name and so on. After all this task was done, the whole data was passed in Elasticsearch where Elasticsearch did the indexing part and then Kibana fetched that data for visualization. Now I will call Shankul for demonstration. This is basically the d-space framework which we were talking about. So this is the recent submission tab which I have implemented. So like these are the six, the latest submission of the courses which have been implemented. We can just redirect it from clicking over here. You can see this is the tag cloud which I have talked about, means this programming language education. Basically in this tag cloud information, we are seeing that whatever the different subject is being used most of the time is talking about the frequency only. So this programming language education would have more frequency in all of the courses. So we can just click over here and it will redirect it to the whatever the courses this subject is talking about. So this is basically the tag cloud and this is the basically discovery portion we are talking about. We can see that there are different courses like we can go into the computer science course. I mean we can see that in the computer science course we have also recent submission. This is different from the home page. It will talk about only the recent submission for this computer science section domain only. We can see there are all those courses which have been implemented recently. We can see this is the course which has been implemented and we will go and we see this is the PDF and this is the iframe tag we will talk about. We can see that this PDF into the dspace only and this is the we can see the YouTube videos. This is the it is streaming live from the YouTube and we can see this and we can go on to any of the portions. So this is the basically the we were explaining and this is and this is basically the dashboard I was talking about. I have like in the earlier I was just implemented this basic electronics course into my dashboard and I can just redirect it to this course only from this dashboard only. So this was the basic demo and any questions. So thank you. Thank you.