Today, the Internet contains a massive amount of information, ranging from social media posts to digital news to customer reviews, etc. Most likely, whatever problem you are trying to tackle, the Internet is going to have related data that is going to help you build a solution. However, in order to fully utilize the data on the Internet, we need to develop a way to collect it.
A popular method of data collection is web scraping. Web scraping attempts to extract useful data from websites. …
In our current age, there is plenty of data from various sources, especially textual data. In a data-driven generation, technologies such as Machine Learning and Natural Language Processing fully leveraged the power of natural language data to analyze and extract interesting insights that were not possible before. In the process of analyzing textual data, it is almost a necessary step to preprocess it before feeding it to a model. In the preprocessing step, it might be useful to search for a specific pattern within an input text.
Here’s where regular expression comes in! Regular expression attempts to find whether a…
For many, pursuing a graduate degree like Master’s or Ph.D. is a great path to enhance your skills and prepare better for the workforce. However, it is crucial to understand the differences between the two and know which one you want to pursue. If you are confused about which one to do, feel free to take a look at my previous post explaining Master’s vs Ph.D. and my decision regarding it.
If you have decided to go with Master’s, in particular AI or Data Science, you might be planning out the programs you want to apply. As somebody who just…
In my previous post, I have described my experiences in getting an undergraduate degree in Computer Science. I also promised that I am going to let you guys know what I’m going to do next. So… [drumroll]… I will be heading to Carnegie Mellon to study for a Master's in Computational Data Science!
For me, the decision to do a Master's was not an easy one. In fact, I always thought I’m going to do a Ph.D. …
For those of you who have followed me for a while or if you are a friend of mine reading this, you might know that I am currently studying at the Hong Kong University of Science and Technology. After three years of effort, I am pleased to announce that I just graduated, with a Bachelor's degree of Engineering in Computer Science!
Sentiment Analysis has been a very popular task since the dawn of Natural Language Processing (NLP). It belongs to a subtask or application of text classification, where sentiments or subjective information from different texts are extracted and identified. Today, many businesses around the world use sentiment analysis to understand more deeply their customers and clients by analyzing sentiments across different target groups. It also has wide applications in different sources of information, including product reviews, online social media, survey feedback, etc.
In this article, we will show you how to implement sentiment analysis quickly and effectively using the Transformers library…
Conversational systems, or dialogue systems, have garnered huge interest in the modern Natural Language Processing (NLP) community. It is simply exciting to see how closely bots can mimic our thoughts, logic, and emotions as shown from their language. Today, we know that there are digital assistants right at the palm of our hands in our smartphones, such as Apple Siri, Google Assistant, and Microsoft Cortana. They are all able to listen and respond to the user’s language although not perfect.
Text Generation is one of the most exciting applications of Natural Language Processing (NLP) in recent years. Most of us have probably heard of GPT-3, a powerful language model that can possibly generate close to human-level texts. However, models like these are extremely difficult to train because of their heavy size, so pretrained models are usually preferred where applicable.
Question answering is a task in information retrieval and Natural Language Processing (NLP) that investigates software that can answer questions asked by humans in natural language. In Extractive Question Answering, a context is provided so that the model can refer to it and make predictions on where the answer lies within the passage.
In this article, we will show you how to implement question answering using pretrained models provided by the Huggingface Transformers library. Since the implementation is really straightforward, you can get your question answering system to work fast within minutes!
Now, let’s get started!
Translation, or more formally, machine translation, is one of the most popular tasks in Natural Language Processing (NLP) that deals with translating from one language to another. In the early days, translation is initially done by simply substituting words in one language to words in another. However, doing that does not yield good results since languages are fundamentally different so a higher level of understanding (e.g. phrases/sentences) is needed. With the advent of deep learning, modern software now adopts statistical and neural techniques, which are proven to be more effective when doing translation.