China
Africa
Explore chapters and articles related to this topic
Artificial Intelligence-Based Categorization of Healthcare Text
Published in Ankan Bhattacharya, Bappadittya Roy, Samarendra Nath Sur, Saurav Mallik, Subhasis Dasgupta, Internet of Things and Data Mining for Modern Engineering and Healthcare Applications, 2023
Global Vectors for Word Representation (GLOVE) is essentially an unsupervised algorithm. To construct word embeddings from a corpus, GLOVE assembles a comprehensive word co-occurrence matrix [5]. GLOVE carries out dimensionality reduction on the co-occurrence information to learn vectors. GLOVE is a count-based model and easier to parallelize compared to the Word2Vec method.
A Deep Learning Approach to Classify the Causes of Depression from Reddit Posts
Published in Roshani Raut, Salah-ddine Krit, Prasenjit Chatterjee, Machine Vision for Industry 4.0, 2022
Ankita Biswas, Arion Mitra, Ananya Ghosh, Namrata Das, Nikita Ghosh, Ahona Ghosh
GloVe stands for “Global Vectors” [Pennington, J., Socher, R., & Manning, C.D. 2014]. GloVe captures both global and local statistics of a corpus, which means not only does GloVe take into consideration the semantics of surrounding words but also the global context. Figure 7.9 illustrates the embedding vector.
Hybrid Attention-based Approach for Arabic Paraphrase Detection
Published in Applied Artificial Intelligence, 2021
GloVe is employed for efficiently capturing the contextual relationships between words (Mahmoud and Zrigui 2021b). It learns semantics and grammar information, taking into account the context of words and the information of the global corpus. Formally, GloVe builds a matrix of word–word co-occurrences by estimating the probability of appearance of a word in the context of another word . It is based on an objective function to produce vectors with a fixed dimension according to a vocabulary size , scalar biases and and a weighting co-occurrence function for rare and frequent words. It is defined as follows in Equation (1):
Keyphrase Extraction Using Enhanced Word and Document Embedding
Published in IETE Journal of Research, 2022
Fahd Saleh Alotaibi, Saurabh Sharma, Vishal Gupta, Savita Gupta
In Ref. [40], the author proposed a local word vectors-based key-phrase extraction technique. The GloVe embedding is used to generate word vectors. This vector representation encodes the contribution of words as eloquent writing. The fundamental idea here is to provide the neighborhood of each word and its local contexts. The reference vector is calculated by averaging the local word vectors of the title and abstract of a document. The author in Ref. [41] proposed a novel optimized approach for keyword extraction by combining the word vector and Text-Rank algorithm. For this, the Word2vec training word vector has been used to obtain the semantic relationship between the words and the doc2vec training vector used to train the paragraph vectors. In Ref. [42], the author proposed a novel topic prediction algorithm that extracts topics from documents using word embeddings. Two novel approaches are designed here. (i) The first approach discusses the standard KNN method which uses a novel alignment-based distance metric. This metric measures the distance between the target sample and all labeled data points. (ii) The second approach highlights the alignment-based threshold classifier, which only requires a predefined threshold value and positively labeled data to measure the distance from a target unknown sample to positively labeled data. In Ref. [43], the author proposed various statistical, advanced and external knowledge-based features to score candidate phrases. Further two different automatic feature selection schemes are adopted here to filter out the best features for a document. The proposed scheme has the following drawbacks: (i) too many over-generated key phrases (ii) most of the features are not applicable for short-length documents.
Detection of Hate Speech using BERT and Hate Speech Word Embedding with Deep Model
Published in Applied Artificial Intelligence, 2023
Hind Saleh, Areej Alhothali, Kawthar Moria
GloVe (Global Vectors for Word Representation) is another popular word embedding model (Pennington, Socher, and Manning 2014). GloVe learns embeddings using an unsupervised learning algorithm that is trained on a corpus to create the distributional feature vectors. During the learning process, a statistics-based matrix is built to represent the word-to-word co-occurrence of the corpus. The main difference between GloVe and Word2Vec is in the learning process, Word2Vec is a prediction-based model, while GloVe is a count-based model. The GloVe is learned from Wikipedia, web data, and Twitter and it has models with different vector dimensions.