In recent years, there has been a growing focus on the veracity of datasets. This concern has raised important questions such as: is the data appropriate for answering the research questions or hypotheses, is the data biased or harmful in any way, is the data quality data, and is there a sufficient understanding of the data for it to be used appropriately? We reviewed 46 papers from Google Scholar, IEEE, and ACM, and found that the majority of authors provide only a basic discussion of the dataset used in the research and do not address important issues such as potential bias or data that requires special attention. Following the work of Bender and Friedman, we propose a data statement framework specifically targeted to computational financial research to provide critical information to users. We also provide a completed data statement for published work as an example. This tool will help researchers provide users and stakeholders a better understanding of what comprises the data and provide an overview of what considerations were made in its creation. This will also help address any potential bias, errors or problems, and data that could be considered misleading in the context of the research.

Bitcoin alongside other cryptocurrencies became one of the largest trends recently, due to its redefinition of the concept of money, and its price fluctuation. Especially on the social media, people keep discussing Bitcoin topics, consulting, and advising about cryptocurrency trading. This paper explores the relationship between Twitter feed on Bitcoin and sentiment analysis of it, comparing and evaluating different data mining classifiers and deep learning methods that might help in better sentiment classification of Bitcoin tweets, the study uses different language modeling approaches, such as tweet embedding and N-Gram modeling. We also evaluate the quality of automated sentiment classification in comparison to manually assigned sentiment labeling. The results show that the manual approach gives significantly better results in some datasets, and superior performance of MLP, WiSARD and decision tree methods. On the other hand, R-Auto Tweets Sentiment (RATS) gives more stable performance overall datasets. using time-series, we found partial correlation between Bitcoin price fluctuation and sentiment class accuracy fluctuations using different machine learning algorithms.

Stock prediction has been a popular research topic and researchers have done a lot of work in this field. Due to its stochastic nature, predicting the future stock market remains a very difficult problem. This paper studies the application of attention-based LSTM deep neural network in future stock market movement prediction. We also build stock aggregate dataset and individual dataset including stock history data, financial tweets sentiment and technical indicators in the US stock market. The experiment studies the time sensitivity of finance tweet sentiment and methods of collective sentiment calculation. This paper also experiments on conventional LSTM and attention-based LSTM for performance comparison. We find the finance tweets that are posted from market closure to market open in the next day has more predictive power on next day stock movement. The weighted sentiment on max follower on StockTwits also outperforms other methods. In our experiment, the result on our individual stock dataset shows a similar pattern like normal distribution.

Extracting key terms from technical documents allows us to write effective documentation that is specific and clear, with minimum ambiguity and confusion caused by nearly synonymous but different terms. For instance, in order to avoid confusion, the same object should not be referred to by two different names (e.g. "hydraulic oil filter"). In the modern world of commerce, clear terminology is the hallmark of successful RFPs (Requests for Proposal) and is therefore a key to the growth of competitive organizations. While Automatic Term Extraction (ATE) is a well-developed area of study, its applications in the technical domain have been sparse and constrained to certain narrow areas such as the biomedical research domain. We present a method for Automatic Term Extraction (ATE) for the technical domain based on the use of part-of-speech features and common words information. The method is evaluated on a C programming language reference manual as well as a manual of aircraft maintenance guidelines, and has shown comparable or better results to the reported state of the art results.

We present a novel marketing method for consumer trend detection from online user generated content, which is motivated by the gap identified in the market research literature. The existing approaches for trend analysis are generally based on rating of trends by industry experts through survey questionnaires, interviews, or similar. These methods proved to be inherently costly and often suffer from bias. Our approach is based on the use of information extraction techniques for identification of trends in large aggregations of social media data. It is cost-effective method that reduces the possibility of errors associated with the design of the sample and the research instrument. The effectiveness of the approach is demonstrated in the experiment performed on restaurant review data. The accuracy of the results is at the level of current approaches for both, information extraction and market research.

The problem of Language Identification (LID) has been present in the Natural Language Processing domain for a relatively long period of time. There is a number of approaches based on statistical methods used for this particular task, and lately AI methods with the revival of neural network techniques. Some of the solutions described and implemented in the past show good performance, but texts that were processed were usually clean in terms of grammatical correctness and homogeneity. This paper explores and discusses LID in short and noisy messages written in similar languages, which is a non-trivial task, especially for very related languages. The experimentation methodology in the paper is based on the algorithms such as SVM, Naïve Bayes variants, Random Forest and Logistic Regression. In addition, we explore a novel distance based classification method - Common N-Grams (CNG). Finally, we explored whether Wikipedia as an additional training data source can improve a classifier performance.

Extracting morphemes from words is a nontrivial task. Rule based stemming approaches such as Porter's algorithm have encountered some success, however they are restricted by their ability to identify a limited number of affixes and are language dependent. When dealing with languages with many affixes, rule based approaches generally require many more rules to deal with all the possible word forms. Deriving these rules requires a larger effort on the part of linguists and in some instances can be simply impractical. We propose an unsupervised ngram based approach, named Swordfish. Using ngram probabilities in the corpus, possible morphemes are identified. We look at two possible methods for identifying candidate morphemes, one using joint probabilities between two ngrams, and the second based on log odds between prefix probabilities. Initial results indicate the joint probability approach to be better for English while the prefix ratio approach is better for Finnish and Turkish.

We propose a novel method for document clustering using character N-grams. In the traditional vector-space model, the documents are represented as vectors, in which each dimension corresponds to a word. We propose a document representation based on the most frequent character N-grams, with window size of up to 10 characters. We derive a new distance measure, which produces uniformly better results when compared to the word-based and term-based methods. The result becomes more significant in the light of the robustness of the N-gram method with no language-dependent preprocessing. Experiments on the performance of a clustering algorithm on a variety of test document corpora demonstrate that the N-gram representation with n=3 outperforms both word and term representations. The comparison between word and term representations depends on the data set and the selected dimensionality.

We present a generic natural language processing (NLP) architecture, acronym QTIL, based on a system of cooperating multiple agents (Q/A, T, I, and L agents) which can be used in any information system incorporating Internet information retrieval. We then introduce a hybrid multi-agent system (MAS) architecture, acronym QTIP, for the privacy domain through integrating the PeCAN (personal context agent networking) and QTIL MAS architectures. There are two areas where NLP is used: in the user-MAS interaction and in the process of resource indexing and matching. These two areas map to the Q/A-agent and to the I-agents. We propose using a lightweight head-driven phrase structure grammar (HPSG) natural language method for the Q architectural layers and qualitatively justify its applicability. We provide an example of employing the HPSG formalism for information retrieval using natural language capability via privacy Web services in one instantiation of the QTIP architecture. Independent preliminary results for HPSG on the Q level show that our approaches for enhancing the usability of PET tools are promising.