Summarize essential information from research meta-data with text mining methods in a few lines of Python code — By: Petr Koráb (Zeppelin University in Friedrichshafen, Germany; Lentiamo, Prague) and Jarko Fidrmuc (Zeppelin University in Friedrichshafen, Germany) Introduction Research publication meta-data (article titles, publication dates, keywords) provide valuable insights into how the whole field develops over time. Text mining with appropriate techniques helps discover which concepts, theories, or models were…

FinVADER is a Python library extending the popular VADER sentiment classifier with two financial lexicons. Learn more in this tutorial. — Introduction Sentiment analysis uses various methods to evaluate sentiment from text data. Deep learning (transformer-based specifically) models like Google’s BERT or Meta’s RoBERTa reach classification accuracy far beyond 94 %. The performance of sentiment classifiers strongly depends on the nature of the data. Some fields, like finance and economics, use specific…

Widely used VADER sentiment classifier uses a general-language lexicon to classify language expressed on social media. This article shows how extending the core lexicon with two domain-specific lexicons improves classification accuracy from 58 % to 69 % on Financial PhraseBank data. Introduction VADER (Valence Aware Dictionary and sEntiment Reasoner) classifier is…

Product reviews are an excellent source of information for qualified management decisions. Learn more about the right text mining techniques. — Introduction Happy customers drive company growth. The five-word sentence explains everything about why we do our best to maximize customer satisfaction. Product reviews are one of the major data sources that large companies like Amazon and Apple, middle-sized exporters including Lentiamo, and local companies running their Facebook pages collect. Reviews are…

Arabica now offers a structural break and sentiment analysis module to enrich time-series text mining — Introduction Text data contains lots of qualitative information, which can be quantified with various methods, including sentiment analysis techniques. These models are used to identify, extract and quantify emotions from text data and have wide use in business and academic research. Since the text is often recorded on a time-series basis…

Have you ever thought about how sentiment from text data can be used as a regressor in time-series models? — Introduction Text data offer qualitative information that can be quantified, aggregated, and used as a variable in time-series models. Simple methods of text data representation, such as one-hot encoding of categorical variables and word n-grams, have been used since NLP’s early beginnings. Over time, more complex methods, including the Bag-of-words model…

Arabica now offers unigram, bigram, and trigram word cloud, heatmap, and line chart to further accelerate time-series text data analysis — Introduction Arabica is a python library for exploratory text data analysis focusing on text from a time-series perspective. It reflects the empirical reality that many text datasets are collected as repeated observations over time. Time series text data include newspaper article headlines, research article abstracts and metadata, product reviews, social network…

Arabica now has easy-to-follow documentation, including coding examples for descriptive and time-series text data analysis — Arabica is a Python library for exploratory data analysis specifically designed for time-series text data. It reflects the reality that many text datasets are currently collected as repeated observations over time (product reviews, communication on social media, article headlines, etc.). It makes exploratory analysis of these datasets simple by providing:

Contour plots are simple and very useful graphics for word embedding visualization. This end-to-end tutorial uses IMDb data to illustrate coding in Python. — Introduction Text data vectorization is a necessary step in modern Natural Language Processing (NLP). The underlying concept of word embeddings has been popularised by two Word2Vec models developed by Mikolov et al. (2013a) and Mikolov et al. (2013b). …