Student success, a major focus in higher education, in part, requires students to remain actively engaged in the required coursework. Identifying student disengagement, when a student stops completing coursework, at scale has been a continuing challenge for higher education due to the heterogeneity of traditional college courses. This research uses data from Connect by McGraw-Hill Education, a widely used online learning tool, to build a classifier to identify learning tool disengagement at scale. This classifier was trained and tested on four years of historical data, representing 4.5 million students in 175,000 courses, across 256 disciplines. Results show that the classifier is effective in identifying disengagement within the online learning tool against baselines, across time, and within and across disciplines. The classifier was also effective in identifying students at risk of disengaging from Connect and then earning unsuccessful grades in a pilot course for which the assignments in Connect were worth a relatively small portion of the overall course grade. Because Connect is widely used, this classifier is positioned to be a good tool for instructors and institutions to identify students at risk for disengagement from coursework. Instructors and institutions can use this information to design and implement interventions to improve engagement and improve student success at the institution in key courses.

Educational practitioners and policy makers require evidence supporting claims about educational efficacy. Evidence is often found using causal relationships between education inputs and student learning outcomes. Causal inference covers a wide range of topics in education research, including efficacy studies to prove if a new policy, software, curriculum or intervention is effective in improving student learning outcomes. Randomized controlled trials (RCT) are considered a gold standard method to demonstrate causality. However, these studies are expensive, timely and costly, as well as not being ethical to conduct in many educational contexts. Causality can also be deducted purely from observational data. In this tutorial, we will review methodologies for estimating the causal effects of education inputs on student learning outcomes using observational data. This is an inherently complex task due to many hidden variables and their interrelationships in educational research. In this tutorial, we discuss causal inference in the context of educational research with big data. This is the first tutorial of its kind at Learning Analytics and Knowledge Conference (LAK) that provides a hands-on experience with Python and Apache Spark as a practical tool for educational researchers to conduct causal inference. As a prerequisite, attendees are required to have familiarity with Python.

In this paper, we describe a scalable learning analytics platform which runs generalized analytics models on educational data in parallel. As a proof of concept, we use this platform as a base for an end-to-end automated writing feedback system. The system allows students to view feedback on their writing in near real-time, edit their writing based on the feedback provided, and observe the progression of their performance over time. Providing students with detailed feedback is an important part of improving writing skills and an essential component towards solving Bloom's "two sigma" problem in education. We evaluate the effectiveness of the feedback for students with an ongoing pilot study with 800 students who are using the learning analytics platform in a college English course.

The growth in digital camera usage combined with a worldly abundance of text has translated to a rich new era for a classic problem of pattern recognition, reading. While traditional document processing often faces challenges such as unusual fonts, noise, and unconstrained lexicons, scene text reading amplifies these challenges and introduces new ones such as motion blur, curved layouts, perspective projection, and occlusion among others. Reading scene text is a complex problem involving many details that must be handled effectively for robust, accurate results. In this work, we describe and evaluate a reading system that combines several pieces, using probabilistic methods for coarsely binarizing a given text region, identifying baselines, and jointly performing word and character segmentation during the recognition process. By using scene context to recognize several words together in a line of text, our system gives state-of-the-art performance on three difficult benchmark data sets.

This paper presents a system for open-vocabulary text recognition in images of natural scenes. First, we describe a novel technique for text segmentation that models smooth color changes across images. We combine this with a recognition component based on a conditional random field with histogram of oriented gradients descriptors and incorporate language information from a lexicon to improve recognition performance. Many existing techniques for this problem use language information from a standard lexicon, but these may not include many of the words found in images of the environment, such as storefront signs and street signs. We avoid this limitation by incorporating language information from a large web-based lexicon of around 13.5 million words. This lexicon contains words encountered during a crawl of the web, so it is likely to contain proper nouns, like business names and street names. We show that our text segmentation method allows for better recognition performance than the current state-of-the-art text segmentation method. We also evaluate this full system on two standard data sets, ICDAR 2003 and ICDAR 2011, and show an increase in word recognition performance compared to the current state-of-the-art methods.

This paper presents a new language model for text recognition in natural images. Many existing techniques incorporate n-gram information as an additional source of information. One problem is that some n-grams are very uncommon, but will still appear in a word across a syllable boundary. These words are given a low probability under an n-gram model. To overcome this problem, we introduce a probabilistic syllable model that uses a probabilistic context-free grammar to generate recognized word labels that are consistent with syllables. In other words, labels generated by this model are pronounceable. This is important for scene text recognition where text often includes proper nouns and standard dictionary information cannot be a useful resource. We show that this language model leads to increased recognition accuracy over a big ram model and discuss the benefits over a dictionary model.

Recognizing text in natural photographs that contain specular highlights and focal blur is a challenging problem. In this paper we describe a new text segmentation method based on inverse rendering, i.e. decomposing an input image into basic rendering elements. Our technique uses iterative optimization to solve the rendering parameters, including light source, material properties (e.g. diffuse/specular reflectance and shininess) as well as blur kernel size. We combine our segmentation method with a recognition component and show that by accounting for the rendering parameters, our approach achieves higher text recognition accuracy than previous work, particularly in the presence of color changes and image blur. In addition, the derived rendering parameters can be used to synthesize new text images that imitate the appearance of an existing image.

The recognition of text in everyday scenes is made difficult by viewing conditions, unusual fonts, and lack of linguistic context. Most methods integrate a priori appearance information and some sort of hard or soft constraint on the allowable strings. Weinman and Learned-Miller [14] showed that the similarity among characters, as a supplement to the appearance of the characters with respect to a model, could be used to improve scene text recognition. In this work, we make further improvements to scene text recognition by taking a novel approach to the incorporation of similarity. In particular, we train a similarity expert that learns to classify each pair of characters as equivalent or not. After removing logical inconsistencies in an equivalence graph, we formulate the search for the maximum likelihood interpretation of a sign as an integer program. We incorporate the equivalence information as constraints in the integer program and build an optimization criterion out of appearance features and character bigrams. Finally, we take the optimal solution from the integer program, and compare all “nearby” solutions using a probability model for strings derived from search engine queries. We demonstrate word error reductions of more than 30% relative to previous methods on the same data set.