This paper explores three simple data manipulation techniques (synthesis, augmentation, curriculum) for improving abstractive summarization models without the need for any additional data. We introduce a method of data synthesis with paraphrasing, a data augmentation technique with sample mixing, and curriculum learning with two new difficulty…

This paper explores the effect of using multitask learning for abstractive summarization in the context of small training corpora. In particular, we incorporate four different tasks (extractive summarization, language modeling, concept detection, and paraphrase detection) both individually and in combination, with the goal of enhancing the target…

Improving educational practice through reflection is one important focus of teacher professional development approaches. However, such teacher reflection operates under practical classroom constraints that make it happen infrequently, including the reliance on disruptive peer/supervisor observations or recordings. This article describes three…

This paper provides an alternative way of document representation by treating topic probabilities as a vector representation for words and representing a document as a combination of the word vectors. A comparison on summary data shows that this representation is more effective in document classification. [This paper was published in:…

Recordings of human interaction data can be organized into temporal representations with different affordances. We use audio data of a learning-related discussion analyzed for its low-level emotional indicators and divided into four phases, each characterized by an overarching emotion. After arguing for the relevance of emotion to learning, we…

The increasing amount of available subjective text data in internet such as product reviews, movie critiques and social media comments provides golden opportunities for information retrieval researchers to extract useful information out of such datasets. Topic modeling and sentiment analysis are two widely researched fields that separately try to…

Annotations play an increasingly crucial role in scientific exploration and discovery, as the amount of data and the level of collaboration among scientists increases. There are many systems today focusing on annotation management, querying, and propagation. Although all such systems are implemented to take user input (i.e., the annotations…

Information Product Maps are visual diagrams used to represent the inputs, processing, and outputs of data within an Information Manufacturing System. A data unit, drawn as an edge, symbolizes a grouping of raw data as it travels through this system. Processes, drawn as vertices, transform each data unit input into various forms prior to delivery…

This teaching case discusses the analysis of an electronic voting system. The development of the case was motivated by research into information security and management, but as it includes procedural aspects, organizational structure and personnel, it is a suitable basis for all aspects of systems analysis, planning and design tasks. The material…

The 11th International Conference on Educational Data Mining (EDM 2018) is held under the auspices of the International Educational Data Mining Society at the Templeton Landing in Buffalo, New York. This year's EDM conference was highly competitive, with 145 long and short paper submissions. Of these, 23 were accepted as full papers and 37…

The translation into French of the Encoded Archival Context (EAC) DTD tag library has been in progress for a few months. It is being carried out by a group of experts gathered by AFNOR, the French national standards agency. The main goal of this group is to foster the interoperability of authority data between archives, libraries and museums, and…