Martin Malchow

Lecture Butler: Teaching Reasonable Lectures from a Lecture Video Archive

Lecture video archives offer a large variety of lecture recordings in different topics. Naturally, topics are described superficially, easily or detailed in different lectures. Users interested in certain topics have problems finding lectures describing a topic chronology from basic lectures to more detailed difficult lectures. The Lecture Butler is going to automatically offer e-learning students lectures for the topics of interest in chronological playlists. The approach is finding lecture information using title, description, OCR and ASR data. This data is indexed and searched by an in-memory database to fulfill the speed requirements for playlist creation. In the search results lectures are going to be ordered by lecture occurrence in the university semester time schedule or by given lecture level of difficulty. As a result students can automatically create playlists for their topic of interest in sequence of the lecture level. Hence, students are not overstrained by lectures when they start with basic lectures first. Basic lectures provide information to understand more complex lectures. The research shows that an automatic approach by adding the level of difficulty or university semester time table is going to show reasonable playlists to find topics of interest. This solves the main problem students encounter when they try to learn a topic step-by-step using recorded lectures. The approach will support and motivate students using e-learning opportunities.

Enhance Lecture Archive Search with OCR Slide Detection and In-Memory Database Technology

On the Web there are a lot of frequently used video lecture archives which have grown up fast during the last couple of years. This fact led to a lot of lecture recordings which include knowledge for a variety of subjects. The typical way of searching these videos is by title and description. Unfortunately, not all important keywords and facts are mentioned in the title or description if they are available. Furthermore, there is no possibility to analyze how important those detected keywords are for the whole video. Another lecture archive specific virtue is that every regular university lecture is repeated yearly. Normally this will lead to duplicate lecture recordings. In search results doubling is disturbing for students when they want to watch the most recent lectures from the search result. This paper deals with the idea to resolve these problems by analyzing the recorded lecture slides with Optical Character Recognition (OCR). In addition to the name and description the OCR data will be used for a full text analysis to create an index for the lecture archive search. Furthermore, a fuzzy search is introduced. This will solve the issue of misspelled search requests and OCR detection defects. Additionally, this paper deals with the performance issues of a full text search with an in-memory database, issues in OCR detection, handling duplicate recordings of lectures repeated every year. Finally, an evaluation of the search performance in comparison with other database ideas besides the in-memory database is performed. Additionally, a user acceptability survey for the search results to increase the learning experience on lecture archives was performed. As a result, this paper shows how to handle the big amount of OCR data for a full text live search performed on an in-memory database in reasonable time. During this search a fuzzy search is performed additionally to resolve spelling mistakes and OCR detection problems. In conclusion this paper shows a solution for an enhanced video lecture archive search that supports students in online research processes and enhances their learning experience.

Enhance Teleteaching Videos with Semantic Technologies

Teleteaching systems have existed for more than a decade now. During that time, thousands of lectures have been recorded. These recordings are usually made available as single or dual stream videos. In the case of dual stream one video shows the speaker and the other one the slides. However, the content of the slides and the speaker’s talk are not accessible through automatic search functionality provided by the e-lecture platform hosting the videos. In order to change this situation optical character recognition (OCR), automatic speech recognition (ASR) and common methods of the semantic web are used to analyze the video content and its semantics. Nevertheless, the semantic information found has to be made accessible to the user. Therefore, a novel HTML5 video player is introduced to enable the students to utilize the found semantic information while watching the video. This will enhance and simplify their online learning or research process.

Couch Learning Mode: Multiple-Video Lecture Playlist Selection out of a Lecture Video Archive for E-learning Students

During a video recorded university class students have to watch several hours of video content. This can easily add up to several days of video content during a semester. Naturally, not all 90 minutes of a typical lecture are relevant for the exam. When the semester ends with a final exam students have to study more intensively the important parts of all the lectures. To simplify the learning process and design it to be more efficient we have introduced the Couch Learning Mode in our lecture video archive. With this approach students can create custom playlists out of the video lecture archive with a time frame for every selected video. Finally, students can lean back and watch all relevant video parts consecutively for the exam without being interrupted. Additionally, the students can share their playlists with other students or they can use the video search to watch all relevant lecture videos about a topic. This approach uses playlists and HTML5 technologies to realize the consecutive video playback. Furthermore, the powerful Lecture Butler search engine is used to find worthwhile video parts for certain topics. Our approach shows that we have more satisfied students using the manual playlist creation to view reasonable parts for an exam. Finally, students are keen on watching the top search results showing reasonable parts of lectures for a topic of interest. The Couch Learning Mode supports and motivates students to learn with video lectures for an exam and daily life.