Michael Bloodgood

Luke Kurlandski, Michael Bloodgood

Active learning is an increasingly important branch of machine learning and a powerful technique for natural language processing. The main advantage of active learning is its potential to reduce the amount of labeled data needed to learn high-performing models. A vital aspect of an effective active learning algorithm is the determination of when to stop obtaining additional labeled data. Several leading state-of-the-art stopping methods use a stop set to help make this decision. However, there has been relatively less attention given to the choice of stop set than to the stopping algorithms that are applied on the stop set. Different choices of stop sets can lead to significant differences in stopping method performance. We investigate the impact of different stop set choices on different stopping methods. This paper shows the choice of the stop set can have a significant impact on the performance of stopping methods and the impact is different for stability-based methods from that on confidence-based methods. Furthermore, the unbiased representative stop sets suggested by original authors of methods work better than the systematically biased stop sets used in recently published work, and stopping methods based on stabilizing predictions have stronger performance than confidence-based stopping methods when unbiased representative stop sets are used. We provide the largest quantity of experimental results on the impact of stop sets to date. The findings are important for helping to illuminate the impact of this important aspect of stopping methods that has been under-considered in recently published work and that can have a large practical impact on the performance of stopping methods for important semantic computing applications such as technology assisted review and text classification more broadly.

Thomas Orth, Michael Bloodgood

When creating text classification systems, one of the major bottlenecks is the annotation of training data. Active learning has been proposed to address this bottleneck using stopping methods to minimize the cost of data annotation. An important capability for improving the utility of stopping methods is to effectively forecast the performance of the text classification models. Forecasting can be done through the use of logarithmic models regressed on some portion of the data as learning is progressing. A critical unexplored question is what portion of the data is needed for accurate forecasting. There is a tension, where it is desirable to use less data so that the forecast can be made earlier, which is more useful, versus it being desirable to use more data, so that the forecast can be more accurate. We find that when using active learning it is even more important to generate forecasts earlier so as to make them more useful and not waste annotation effort. We investigate the difference in forecasting difficulty when using accuracy and F-measure as the text classification system performance metrics and we find that F-measure is more difficult to forecast. We conduct experiments on seven text classification datasets in different semantic domains with different characteristics and with three different base machine learning algorithms. We find that forecasting is easiest for decision tree learning, moderate for Support Vector Machines, and most difficult for neural networks.

Garrett Beatty, Ethan Kochis, Michael Bloodgood

Annotation of training data is the major bottleneck in the creation of text classification systems. Active learning is a commonly used technique to reduce the amount of training data one needs to label. A crucial aspect of active learning is determining when to stop labeling data. Three potential sources for informing when to stop active learning are an additional labeled set of data, an unlabeled set of data, and the training data that is labeled during the process of active learning. To date, no one has compared and contrasted the advantages and disadvantages of stopping methods based on these three information sources. We find that stopping methods that use unlabeled data are more effective than methods that use labeled data.

Michael Altschuler, Michael Bloodgood

During active learning, an effective stopping method allows users to limit the number of annotations, which is cost effective. In this paper, a new stopping method called Predicted Change of F Measure will be introduced that attempts to provide the users an estimate of how much performance of the model is changing at each iteration. This stopping method can be applied with any base learner. This method is useful for reducing the data annotation bottleneck encountered when building text classification systems.

Garrett Beatty, Ethan Kochis, Michael Bloodgood

When using active learning, smaller batch sizes are typically more efficient from a learning efficiency perspective. However, in practice due to speed and human annotator considerations, the use of larger batch sizes is necessary. While past work has shown that larger batch sizes decrease learning efficiency from a learning curve perspective, it remains an open question how batch size impacts methods for stopping active learning. We find that large batch sizes degrade the performance of a leading stopping method over and above the degradation that results from reduced learning efficiency. We analyze this degradation and find that it can be mitigated by changing the window size parameter of how many past iterations of learning are taken into account when making the stopping decision. We find that when using larger batch sizes, stopping methods are more effective when smaller window sizes are used.

Michael Bloodgood

This paper investigates and evaluates support vector machine active learning algorithms for use with imbalanced datasets, which commonly arise in many applications such as information extraction applications. Algorithms based on closest-to-hyperplane selection and query-by-committee selection are combined with methods for addressing imbalance such as positive amplification based on prevalence statistics from initial random samples. Three algorithms (ClosestPA, QBagPA, and QBoostPA) are presented and carefully evaluated on datasets for text classification and relation extraction. The ClosestPA algorithm is shown to consistently outperform the other two in a variety of ways and insights are provided as to why this is the case.

Michael Bloodgood, Benjamin Strauss

With the advent of informal electronic communications such as social media, colloquial languages that were historically unwritten are being written for the first time in heavily code-switched environments. We present a method for inducing portions of translation lexicons through the use of expert knowledge in these settings where there are approximately zero resources available other than a language informant, potentially not even large amounts of monolingual data. We investigate inducing a Moroccan Darija-English translation lexicon via French loanwords bridging into English and find that a useful lexicon is induced for human-assisted translation and statistical machine translation.

Michael Bloodgood, Benjamin Strauss

Global constraints and reranking have not been used in cognates detection research to date. We propose methods for using global constraints by performing rescoring of the score matrices produced by state of the art cognates detection systems. Using global constraints to perform rescoring is complementary to state of the art methods for performing cognates detection and results in significant performance improvements beyond current state of the art performance on publicly available datasets with different language pairs and various conditions such as different levels of baseline state of the art performance and different data size conditions, including with more realistic large data size conditions than have been evaluated with in the past.

Alan Mishler, Kevin Wonus, Wendy Chambers et al.

Since the events of the Arab Spring, there has been increased interest in using social media to anticipate social unrest. While efforts have been made toward automated unrest prediction, we focus on filtering the vast volume of tweets to identify tweets relevant to unrest, which can be provided to downstream users for further analysis. We train a supervised classifier that is able to label Arabic language tweets as relevant to unrest with high reliability. We examine the relationship between training data size and performance and investigate ways to optimize the model building process while minimizing cost. We also explore how confidence thresholds can be set to achieve desired levels of performance.

Michael Bloodgood, Benjamin Strauss

Many important forms of data are stored digitally in XML format. Errors can occur in the textual content of the data in the fields of the XML. Fixing these errors manually is time-consuming and expensive, especially for large amounts of data. There is increasing interest in the research, development, and use of automated techniques for assisting with data cleaning. Electronic dictionaries are an important form of data frequently stored in XML format that frequently have errors introduced through a mixture of manual typographical entry errors and optical character recognition errors. In this paper we describe methods for flagging statistical anomalies as likely errors in electronic dictionaries stored in XML format. We describe six systems based on different sources of information. The systems detect errors using various signals in the data including uncommon characters, text length, character-based language models, word-based language models, tied-field length ratios, and tied-field transliteration models. Four of the systems detect errors based on expectations automatically inferred from content within elements of a single field type. We call these single-field systems. Two of the systems detect errors based on correspondence expectations automatically inferred from content within elements of multiple related field types. We call these tied-field systems. For each system, we provide an intuitive analysis of the type of error that it is successful at detecting. Finally, we describe two larger-scale evaluations using crowdsourcing with Amazon's Mechanical Turk platform and using the annotations of a domain expert. The evaluations consistently show that the systems are useful for improving the efficiency with which errors in XML electronic dictionaries can be detected.

Michael Bloodgood, Benjamin Strauss

Translation Memory (TM) systems are one of the most widely used translation technologies. An important part of TM systems is the matching algorithm that determines what translations get retrieved from the bank of available translations to assist the human translator. Although detailed accounts of the matching algorithms used in commercial systems can't be found in the literature, it is widely believed that edit distance algorithms are used. This paper investigates and evaluates the use of several matching algorithms, including the edit distance algorithm that is believed to be at the heart of most modern commercial TM systems. This paper presents results showing how well various matching algorithms correlate with human judgments of helpfulness (collected via crowdsourcing with Amazon's Mechanical Turk). A new algorithm based on weighted n-gram precision that can be adjusted for translator length preferences consistently returns translations judged to be most helpful by translators for multiple domains and language pairs.

Michael Bloodgood, John Grothendieck

Within the natural language processing (NLP) community, active learning has been widely investigated and applied in order to alleviate the annotation bottleneck faced by developers of new NLP systems and technologies. This paper presents the first theoretical analysis of stopping active learning based on stabilizing predictions (SP). The analysis has revealed three elements that are central to the success of the SP method: (1) bounds on Cohen's Kappa agreement between successively trained models impose bounds on differences in F-measure performance of the models; (2) since the stop set does not have to be labeled, it can be made large in practice, helping to guarantee that the results transfer to previously unseen streams of examples at test/application time; and (3) good (low variance) sample estimates of Kappa between successive models can be obtained. Proofs of relationships between the level of Kappa agreement and the difference in performance between consecutive models are presented. Specifically, if the Kappa agreement between two models exceeds a threshold T (where $T>0$), then the difference in F-measure performance between those models is bounded above by $\frac{4(1-T)}{T}$ in all cases. If precision of the positive conjunction of the models is assumed to be $p$, then the bound can be tightened to $\frac{4(1-T)}{(p+1)T}$.

Vinodkumar Prabhakaran, Michael Bloodgood, Mona Diab et al.

We explore training an automatic modality tagger. Modality is the attitude that a speaker might have toward an event or state. One of the main hurdles for training a linguistic tagger is gathering training data. This is particularly problematic for training a tagger for modality because modality triggers are sparse for the overwhelming majority of sentences. We investigate an approach to automatically training a modality tagger where we first gathered sentences based on a high-recall simple rule-based modality tagger and then provided these sentences to Mechanical Turk annotators for further annotation. We used the resulting set of training data to train a precise modality tagger using a multi-class SVM that delivers good performance.

Kathryn Baker, Michael Bloodgood, Bonnie J. Dorr et al.

This paper describes the resource- and system-building efforts of an eight-week Johns Hopkins University Human Language Technology Center of Excellence Summer Camp for Applied Language Exploration (SCALE-2009) on Semantically-Informed Machine Translation (SIMT). We describe a new modality/negation (MN) annotation scheme, the creation of a (publicly available) MN lexicon, and two automated MN taggers that we built using the annotation scheme and lexicon. Our annotation scheme isolates three components of modality and negation: a trigger (a word that conveys modality or negation), a target (an action associated with modality or negation) and a holder (an experiencer of modality). We describe how our MN lexicon was semi-automatically produced and we demonstrate that a structure-based MN tagger results in precision around 86% (depending on genre) for tagging of a standard LDC data set. We apply our MN annotation scheme to statistical machine translation using a syntactic framework that supports the inclusion of semantic annotations. Syntactic tags enriched with semantic annotations are assigned to parse trees in the target-language training texts through a process of tree grafting. While the focus of our work is modality and negation, the tree grafting procedure is general and supports other types of semantic information. We exploit this capability by including named entities, produced by a pre-existing tagger, in addition to the MN elements produced by the taggers described in this paper. The resulting system significantly outperformed a linguistically naive baseline model (Hiero), and reached the highest scores yet reported on the NIST 2009 Urdu-English test set. This finding supports the hypothesis that both syntactic and semantic information can improve translation quality.

Benjamin S. Mericli, Michael Bloodgood

Turkic languages exhibit extensive and diverse etymological relationships among lexical items. These relationships make the Turkic languages promising for exploring automated translation lexicon induction by leveraging cognate and other etymological information. However, due to the extent and diversity of the types of relationships between words, it is not clear how to annotate such information. In this paper, we present a methodology for annotating cognates and etymological origin in Turkic languages. Our method strives to balance the amount of research effort the annotator expends with the utility of the annotations for supporting research on improving automated translation lexicon induction.

John E. Miller, Michael Bloodgood, Manabu Torii et al.

Part-of-speech (POS) tagging is a fundamental component for performing natural language tasks such as parsing, information extraction, and question answering. When POS taggers are trained in one domain and applied in significantly different domains, their performance can degrade dramatically. We present a methodology for rapid adaptation of POS taggers to new domains. Our technique is unsupervised in that a manually annotated corpus for the new domain is not necessary. We use suffix information gathered from large amounts of raw text as well as orthographic information to increase the lexical coverage. We present an experiment in the Biological domain where our POS tagger achieves results comparable to POS taggers specifically trained to this domain.

Michael Bloodgood, Peng Ye, Paul Rodrigues et al.

When digitizing a print bilingual dictionary, whether via optical character recognition or manual entry, it is inevitable that errors are introduced into the electronic version that is created. We investigate automating the process of detecting errors in an XML representation of a digitized print dictionary using a hybrid approach that combines rule-based, feature-based, and language model-based methods. We investigate combining methods and show that using random forests is a promising approach. We find that in isolation, unsupervised methods rival the performance of supervised methods. Random forests typically require training data so we investigate how we can apply random forests to combine individual base methods that are themselves unsupervised without requiring large amounts of training data. Experiments reveal empirically that a relatively small amount of data is sufficient and can potentially be further reduced through specific selection criteria.

Paul Rodrigues, David Zajic, David Doermann et al.

Dictionaries are often developed using tools that save to Extensible Markup Language (XML)-based standards. These standards often allow high-level repeating elements to represent lexical entries, and utilize descendants of these repeating elements to represent the structure within each lexical entry, in the form of an XML tree. In many cases, dictionaries are published that have errors and inconsistencies that are expensive to find manually. This paper discusses a method for dictionary writers to quickly audit structural regularity across entries in a dictionary by using statistical language modeling. The approach learns the patterns of XML nodes that could occur within an XML tree, and then calculates the probability of each XML tree in the dictionary against these patterns to look for entries that diverge from the norm.

David Zajic, Michael Maxwell, David Doermann et al.

We describe a paradigm for combining manual and automatic error correction of noisy structured lexicographic data. Modifications to the structure and underlying text of the lexicographic data are expressed in a simple, interpreted programming language. Dictionary Manipulation Language (DML) commands identify nodes by unique identifiers, and manipulations are performed using simple commands such as create, move, set text, etc. Corrected lexicons are produced by applying sequences of DML commands to the source version of the lexicon. DML commands can be written manually to repair one-off errors or generated automatically to correct recurring problems. We discuss advantages of the paradigm for the task of editing digital bilingual dictionaries.

Michael Bloodgood, Chris Callison-Burch

We explore how to improve machine translation systems by adding more translation data in situations where we already have substantial resources. The main challenge is how to buck the trend of diminishing returns that is commonly encountered. We present an active learning-style data solicitation algorithm to meet this challenge. We test it, gathering annotations via Amazon Mechanical Turk, and find that we get an order of magnitude increase in performance rates of improvement.

Michael Bloodgood, Chris Callison-Burch

Building machine translation (MT) test sets is a relatively expensive task. As MT becomes increasingly desired for more and more language pairs and more and more domains, it becomes necessary to build test sets for each case. In this paper, we investigate using Amazon's Mechanical Turk (MTurk) to make MT test sets cheaply. We find that MTurk can be used to make test sets much cheaper than professionally-produced test sets. More importantly, in experiments with multiple MT systems, we find that the MTurk-produced test sets yield essentially the same conclusions regarding system performance as the professionally-produced test sets yield.

Kathryn Baker, Michael Bloodgood, Bonnie J. Dorr et al.

This paper describes our resource-building results for an eight-week JHU Human Language Technology Center of Excellence Summer Camp for Applied Language Exploration (SCALE-2009) on Semantically-Informed Machine Translation. Specifically, we describe the construction of a modality annotation scheme, a modality lexicon, and two automated modality taggers that were built using the lexicon and annotation scheme. Our annotation scheme is based on identifying three components of modality: a trigger, a target and a holder. We describe how our modality lexicon was produced semi-automatically, expanding from an initial hand-selected list of modality trigger words and phrases. The resulting expanded modality lexicon is being made publicly available. We demonstrate that one tagger---a structure-based tagger---results in precision around 86% (depending on genre) for tagging of a standard LDC data set. In a machine translation application, using the structure-based tagger to annotate English modalities on an English-Urdu training corpus improved the translation quality score for Urdu by 0.3 Bleu points in the face of sparse training data.

Kathryn Baker, Michael Bloodgood, Chris Callison-Burch et al.

We describe a unified and coherent syntactic framework for supporting a semantically-informed syntactic approach to statistical machine translation. Semantically enriched syntactic tags assigned to the target-language training texts improved translation quality. The resulting system significantly outperformed a linguistically naive baseline model (Hiero), and reached the highest scores yet reported on the NIST 2009 Urdu-English translation task. This finding supports the hypothesis (posed by many researchers in the MT community, e.g., in DARPA GALE) that both syntactic and semantic information are critical for improving translation quality---and further demonstrates that large gains can be achieved for low-resource languages with different word order than English.

Michael Bloodgood, K. Vijay-Shanker

A survey of existing methods for stopping active learning (AL) reveals the needs for methods that are: more widely applicable; more aggressive in saving annotations; and more stable across changing datasets. A new method for stopping AL based on stabilizing predictions is presented that addresses these needs. Furthermore, stopping methods are required to handle a broad range of different annotation/performance tradeoff valuations. Despite this, the existing body of work is dominated by conservative methods with little (if any) attention paid to providing users with control over the behavior of stopping methods. The proposed method is shown to fill a gap in the level of aggressiveness available for stopping AL and supports providing users with control over stopping behavior.

Michael Bloodgood, K. Vijay-Shanker

Actively sampled data can have very different characteristics than passively sampled data. Therefore, it's promising to investigate using different inference procedures during AL than are used during passive learning (PL). This general idea is explored in detail for the focused case of AL with cost-weighted SVMs for imbalanced data, a situation that arises for many HLT tasks. The key idea behind the proposed InitPA method for addressing imbalance is to base cost models during AL on an estimate of overall corpus imbalance computed via a small unbiased sample rather than the imbalance in the labeled training data, which is the leading method used during PL.

Michael Bloodgood, K. Vijay-Shanker

There is a broad range of BioNLP tasks for which active learning (AL) can significantly reduce annotation costs and a specific AL algorithm we have developed is particularly effective in reducing annotation costs for these tasks. We have previously developed an AL algorithm called ClosestInitPA that works best with tasks that have the following characteristics: redundancy in training material, burdensome annotation costs, Support Vector Machines (SVMs) work well for the task, and imbalanced datasets (i.e. when set up as a binary classification problem, one class is substantially rarer than the other). Many BioNLP tasks have these characteristics and thus our AL algorithm is a natural approach to apply to BioNLP tasks.