Computer science
- Explain the relationship among data mining, text mining, and sentiment analysis.
Data mining refers to the analysis of large volumes of data with the aim of identification of meaningful patterns. Text mining is the analysis of text data which in an unstructured format and mapping it to a structured format, while sentiment analysis entails the analysis of both data and information and this is based on peoples’ opinions. Therefore, the relationship between the three is that they both involve some form of analysis.
- In your own words, define text mining, and discuss its most popular applications.
Text mining refers to relating structured data to unstructured data with the aim of gaining a better understanding. The most popular application of this is in business intelligence whereby such technique is used in decision making.
- What does it mean to induce structure into text-based data? Discuss the alternative ways of inducing structure into them.
Inducing structure into text-based data is ensuring that the data is free of errors and is accurate. The alternative way of inducing structure into text-data is by editing and proofreading.
- What is the role of NLP in text mining? Discuss the capabilities and limitations of NLP in the context of text mining.
The role of NLP in text mining is to help a machine to read the text. The capability of NLP in the context of text mining is that it can decipher the ambiguities evident in human language. On the other hand, the limitation of NLP in text mining is that it requires a consistent knowledge base.
EXERCISE
Go to kdnuggets.com. Explore the sections on applications as well as software. Find names of at least three additional packages for data mining and text mining.
One of the packages for data mining and text mining is the agents and bots. This package is useful in business intelligence. The various software that are under the agents and bots include artificial life, autonomy, BotXO, LPA Chimera Agent Toolkit along with Web QL from QL2. When it comes to the first software which is artificial life, this develops, markets, and supports intelligent robots from the internet. On the other hand, autonomy is used in the modification of uncategorized information. BotXO software, under the agents and bots application, builds software for the conversational area and this is by working with chatbots, Machine learning, and AI, along with NLU. Last but not least, LPA Chimera Agent Toolkit is a powerful deployment tool while WebQL from QL2 offers an SQL-like syntax for the creation of custom agents along with robots for information extraction from the web along with text sources.
The other packages for data mining and text mining is the anomaly detection, deviation along with fraud detection software. This comprises of machine analytics, penny analytics, GritBot, and Anodot. When it comes to machine analytics, this is an anomaly detection and analytics for machine data and log files. Penny analytics, on the other hand, operates analytics services online where one is supposed to upload files and wait for the results. GritBot helps in the identification of anomalies in data. Anodot offers real-time analytics and also helps in finding outliers for Big data.
Last but not least, there is an estimation and forecasting package. This comes with tools like Cubist, which is used in the estimation of continuous value and is based on a decision tree. The other component under the estimation and forecasting package is the Previa Studio, and this is a specialized interactive environment that gives a definition of the forecasting models.