China
Africa
Explore chapters and articles related to this topic
m-Health: Community-Based Android Application for Medical Services
Published in Adwitiya Sinha, Megha Rathi, Smart Healthcare Systems, 2019
Mahima Narang, Charu Nigam, Nisha Chaurasia
The users can also choose their preference based on the ratings of the health unit calculated by the reviews provided for the hospital. For predicting the rating of responses for queries fired by users, TF-IDF vectorization is used. TF-IDF stands for term frequency–inverse document frequency. First, calculate IDF for each word in the document by performing the logarithmic ratio of the total number of documents to the number of documents in which that word is present. The obtained IDF is multiplied by the total number of that word present in the entire document, which is termed as term frequency. The obtained result is known as TF-IDF vector (Using TF-IDF to Determine Word Relevance in Document Queries). Following includes an example for calculating TF-IDF (Diana, 2016). Our experiment takes into account a collection of four documents given later:
Function-Based Malware Detection Technique for Android
Published in Georgios Kambourakis, Asaf Shabtai, Constantinos Kolias, Dimitrios Damopoulos, Intrusion Detection and Prevention for Mobile Ecosystems, 2017
We apply this class of features from the text categorization domain to our function-based malware detection task. For each function, we calculate its tf-idf measure, which is a well-known measure in the text categorization field, often used as a weighting factor in information retrieval and text mining [40]. The acronym tf-idfis short for term frequency-inverse document frequency. It is a numerical statistic intended to reflect how important a word (i.e., term) is to a document in a collection or corpus. The tf-idf value increases proportionally to the number of times a word appears in the document, but is offset by the frequency of the word in the corpus, which helps to adjust for the fact that some words appear more frequently in general. In addition to the many uses of this measure for search engines as a central tool in scoring and ranking documents, it is also used for classification and malware detection [39].
Relevance in web search
Published in Jun Wu, Rachel Wu, Yuxi Candice Wang, The Beauty of Mathematics in Computer Science, 2018
The concept of TF-IDF is widely recognized as the most important invention in information retrieval, with widespread applications in document classification and other fields. Though famous today, TF-IDF had a rough start. In 1972, Karen Spärck Jones* from the University of Cambridge published “A statistical interpretation of term specificity and its application in retrieval,” in which TF-IDF was first proposed. Unfortunately, she did not thoroughly explain its theoretical underpinnings, including why IDF should be an inverse logarithmic function, rather than a square root, for instance. She also did not investigate further on this topic, so later researchers generally do not cite her essay when referring to TF-IDF (and many do not know of her contribution).
A multi-objective framework for the identification and optimisation of factors affecting cybersecurity in the Industry 4.0 supply chain
Published in International Journal of Production Research, 2023
Mayank Shukla, S.P. Sarmah, Manoj Kumar Tiwari
The assumed risk and threat clusters are identified from sources and listed in Table IV of the appendix. Compared to the phrases, paragraphs, and articles scrapped from authentic web sources, such as purplesec.us (Allen 2021), CSIS (Lewis 2021) quantitative details are evaluated using a numerical static known as TF-IDF. Scores related to each cluster belonging to risk and threat are evaluated with every updated news. TF-IDF is intended to extract the importance of individual words in a cluster or paragraph. As shown in Table 2, the accumulated cluster score as TF-IDF is recalculated for each new entry of information in the data frame, and the potential for loss in the form of a rectangular matrix is represented, as mentioned in Table 2. The computed score signifies the importance of a word or a phrase by evaluating matching and similarity scores from relevant news, articles, phrases, related literature, and published articles.
A Novel Hybrid Machine Learning Model for Analyzing E-Learning Users’ Satisfaction
Published in International Journal of Human–Computer Interaction, 2023
Sulis Sandiwarno, Zhendong Niu, Ally S. Nyamawe
Table 3 depicts the classification results based on TF-IDF, TF-IWF, Word2Vec, GloVe, fastText, and BERT algorithms in terms of average macro-precision (Pre), recall (Rec), and F1-score (F1). From Table 3 we observe that based on TF-IDF algorithm the machine learning classifiers can accurately classify users’ opinions with an average at least 62.64% of F1. Whereas, based on Word2Vec the classifiers can accurately classify opinions with an average at least 68.09% of F1. In addition, El-USD using Word2Vec outperforms TF-IWF algorithm with an average improvement of 2.05, 2.03, and 2% in terms of precision, recall, and F1, respectively. Broadly speaking, TF-IDF has been shown to be effective and slightly improves the classification performance. Additionally, TF-IDF algorithm can obtain good performance on short textual dataset and it can mark the importance of word (Hasan & Ng, 2014). However, the limitation of TF-IDF is that it ignores the importance of Word Frequency (WF) in the text. To address the TF-IDF limitation, Wang et al. proposed TF-IWF algorithm. TF-IWF algorithm influences the WF reciprocal to replace the Inverse Document Frequency (IDF) with Inverse Word Frequency (IWF) (Wang et al., 2008). Although, the TF-IWF algorithm is the conventional feature extraction technique, it can successfully employ document frequencies to compute the normalized frequency in view of a corpus. Therefore, to support in investigating users’ satisfaction in e-learning system based on opinions, the TF-IWF algorithm outperforms TF-IDF feature extraction techniques.
Intelligent Grouping Method of Science and Technology Projects Based on Data Augmentation and SMOTE
Published in Applied Artificial Intelligence, 2022
TF-IDF is a feature weighting technique commonly used in information retrieval and data mining (Kim and Gil 2019; Liu et al. 2018; Zhu et al. 2016b). The key idea of TF-IDF lies in that a word is not trivial to the text when it gets a high frequency in a text. Furthermore, if the word rarely or even does not appear in other texts except for the current text in the text set, the word has a strong ability to distinguish the current text and other texts. TF of TF-IDF is term frequency, which represents the frequency of occurrence of a word in the text. IDF of TF-IDF is the inverse document frequency, which means the lower word frequency is in other texts, the higher the IDF value is accordingly. The calculation formulation of the TF value of the word ti in the text dj is given as follows: