How does 🏹.to work?

How does 🏹.to work?

10.Oct.2021

 

🏹.to is powered by Natural Language Processing (NLP) techniques to look for contextual information in the URL to be shortened. Our 🤖 extracts words from the URL and the destination page metadata and maps them to semantically-related emojis to automagically™ generate a meaningful short link.

 

How does 🏹.to work?

Natural Language Processing (NLP) uses machine learning algorithms that are able to "understand" human speech. We use various NLP tools, specifically Word Embedding using singular value decomposition of word2vec models, which learns relationships between words in documents, then builds relationships between these words into vectors that are used in similarity calculations when mapping terms in URLs or pages to emojis.

 

We use various other NLP techniques including language detection, lemmatization, part-of-speech tagging, chunking etc. On top of this knowledge graph we are able to extract semantic information from the URL and metadata on the destination page. More information about how our 🤖 calculates these relationships will be available soon™ as part of a series of articles aimed at demystifying and opening up our technology and processes to the community.

Now you know!

How does 🏹.to work? We use Natural Language Processing (NLP) techniques to look for contextual information in the URL to be shortened. Our 🤖 extracts words from the URL and metadata on the destination page, then maps them to semantically-related emojis to automagically™ generate a meaningful short link.

 

How does 🏹.to work? Use of "Natural Language Processing" techniques, specifically word embeddings, uses machine learning algorithms that are able to understand human speech. We use these tools, which were trained using singular value decomposition of word2vec models, to extract relationships between words in documents, and build relationships between these words into vectors that are used in similarity calculations when mapping terms in URLs or pages to emojis.

 

We use various other NLP techniques including language detection, lemmatization, part-of-speech tagging, chunking etc. On top of this knowledge graph we are able to extract semantic information from the URL and metadata on the destination page. More information about how our 🤖 calculates these relationships will be available soon™ as part of a series of articles aimed at demystifying and opening up our technology and processes to the community.

Now you know!

How does 🏹.to work? Use of NLTK, specifically word embeddings, uses machine learning algorithms that are able to understand human speech. We use these tools, which were trained using singular value decomposition of word2vec models, to extract relationships between words in documents, then build relationships between these words into vectors that are used in similarity calculations when mapping terms in URLs or pages to emojis

 

We use various other NLP techniques including language detection, lemmatization, part-of-speech tagging, chunking etc. On top of this knowledge graph we are able to extract semantic information from the URL and metadata on the destination page. More information about how our 🤖 calculates these relationships will be available soon™ as part of a series of articles aimed at demystifying and opening up our technology and processes to the community.

Now you know!

How does 🏹.to work? The use of NLTK words embeddings uses machine learning algorithms that are able to understand human speech. We use these tools, which were trained using singular value decomposition of word2vec models, to extract relationships between words in documents, then build relationships between these words into vectors that are used in similarity calculations when mapping terms in URLs or pages to emojis.

 

We use various other NLP techniques including language detection, lemmatization, part-of-speech tagging, chunking etc. On top of this knowledge graph we are able to extract semantic information from the URL and metadata on the destination page. More information about how our 🤖 calculates these relationships will be available soon™ as part of a series of articles aimed at demystifying and opening up our technology and processes to the community.

Now you know! 

 

how does 🏹.to work? The use of NLTK uses machine learning algorithms that are able to understand human speech. We use these tools, which were trained using singular value decomposition of word2vec models, to extract relationships between words in documents, then build relationships between these words into vectors that are used in similarity calculations when mapping terms in URLs or pages to emojis

 

We use various other NLP techniques including language detection, lemmatization, part-of-speech tagging, chunking etc. On top of this knowledge graph we are able to extract semantic information from the URL and metadata on the destination page. More information about how our 🤖 calculates these relationships will be available soon™ as part of a series of articles aimed at demystifying and opening up our technology and processes to the community.

We are social