Semantic text classification: A survey of past and recent advances

Additionally, we consider a weight propagation mechanism that exploits semantic relationships in the concept graph and conveys a spreading activation component. We enrich word2vec embeddings with the resulting semantic vector through concatenation or replacement and apply the semantically augmented word embeddings on the classification task via a DNN. Experimental results over established datasets demonstrate that our approach of semantic augmentation in the input space boosts classification performance significantly, with concatenation offering the best performance. This is accomplished by post-processing the existing word vectors to balance their distance between their original fitted values and their semantic neighbors. The experimental analysis demonstrates the resulting improvements on the embeddings in a multilingual setting, with respect to a variety of semantic-content tasks.

When considering semantics-concerned text mining, we believe that this lack can be filled with the development of good knowledge bases and natural language processing methods specific for these languages. Besides, the analysis of the impact of languages in semantic-concerned text mining is also an interesting open research question. A comparison among semantic aspects of different languages and their impact on the results of text mining techniques would also be interesting. To tackle word polysemy and under/misrepresentations of semantic relationships in the training text, many approaches build embeddings for semantic concepts (or “senses”), instead of words.

Semantic Classification Models

The outcomes demonstrate that the proposed methodology is more prominent than the TF-idf score in ranking the terms with respect to the search query terms. The Pearson correlation coefficient achieved for the semantic similarity model is 0.875. This strategy bears some resemblance to other embedding-based disambiguation methods in the literature.

Hence, it is critical to identify which meaning suits the word depending on its usage. Insights derived from data also help teams detect areas of improvement and make better decisions. For example, you might decide to create a strong knowledge base by identifying the most common customer inquiries. In-Text Classification, our aim is to label the text according to the insights we intend to gain from the textual data. Likewise, the word ‘rock’ may mean ‘a stone‘ or ‘a genre of music‘ – hence, the accurate meaning of the word is highly dependent upon its context and usage in the text. Hence, under Compositional Semantics Analysis, we try to understand how combinations of individual words form the meaning of the text.

Save article to Google Drive

Therefore, they need to be taught the correct interpretation of sentences depending on the context. 1 A simple search for “systematic review” on the Scopus database in June 2016 returned, by subject area, 130,546 Health Sciences documents (125,254 of them for Medicine) and only 5,539 Physical Sciences (1328 of them for Computer Science). The coverage of Scopus publications are balanced between Health Sciences (32% of total Scopus publication) and Physical Sciences (29% of total Scopus publication). Text mining initiatives can get some advantage by using external sources of knowledge. Thesauruses, taxonomies, ontologies, and semantic networks are knowledge sources that are commonly used by the text mining community. Semantic networks is a network whose nodes are concepts that are linked by semantic relations.

However, vector space demonstration of texts usually results in high dimensionality and consequently high sparsity. This is a big difficulty especially when there are numerous class labels but inadequate training data for each of them. Obtaining labeled quality data for training is usually very expensive in real world applications. Accordingly, an accurate text classifier should have the capability of using this semantic information. Uber uses semantic analysis to analyze users’ satisfaction or dissatisfaction levels via social listening.

It supports moderation of users’ comments published on the Polish news portal called Wirtualna Polska. In particular, it aims at finding comments containing offensive words and hate speech. This technique is used separately or can be used along with one of the above methods to gain more valuable insights. For Example, Tagging Twitter mentions by sentiment to get a sense of how customers feel about your product and can identify unhappy customers in real-time.

Search engines can provide more relevant results by understanding user queries better, considering the context and meaning rather than just keywords. This survey investigates the existing and recent advancements in the semantic text classification field and highlights strengths in comparison to the traditional text classification approach. This section presents a summary comparison with respect to a number of key criteria. For example, semantic analysis can generate a repository of the most common customer inquiries and then decide how to address or respond to them. In Natural Language, the meaning of a word may vary as per its usage in sentences and the context of the text. Word Sense Disambiguation involves interpreting the meaning of a word based upon the context of its occurrence in a text.

Semantic Analysis Techniques

However, the participation of users (domain experts) is seldom explored in scientific papers. The difficulty inherent to the evaluation of a method based on user’s interaction is a probable reason for the lack of studies considering this approach. It was surprising to find the high presence of the Chinese language among the studies.

• However, the proposed solutions are normally developed for a specific domain or are language dependent.
• This is a big difficulty especially when there are numerous class labels but inadequate training data for each of them.
• Besides, WordNet can support the computation of semantic similarity [70, 71] and the evaluation of the discovered knowledge [72].

Below, we will denote string literals with a quoted block of text (e.g., “dog”). In many companies, these automated assistants are the first source of contact with customers. The most advanced ones use semantic analysis to understand customer needs and more. For example, the phrase “Time flies like an arrow” can have more than one meaning. If the translator does not use semantic analysis, it may not recognise the proper meaning of the sentence in the given context. Based on them, the classification model can learn to generalise the classification to words that have not previously occurred in the training set.

Contents

Since 2019, Cdiscount has been using a semantic analysis solution to process all of its customer reviews online. This kind of system can detect priority axes of improvement to put in place, based on post-purchase feedback. The company can therefore analyze the satisfaction and dissatisfaction of different consumers through the semantic analysis of its reviews. One of the simplest and most popular methods of finding semantic text analysis meaning in text used in semantic analysis is the so-called Bag-of-Words approach. Thanks to that, we can obtain a numerical vector, which tells us how many times a particular word has appeared in a given text. Despite the fact that the user would have an important role in a real application of text mining methods, there is not much investment on user’s interaction in text mining research studies.

However, it is possible to conduct it in a controlled and well-defined way through a systematic process. Therefore, in semantic analysis with machine learning, computers use Word Sense Disambiguation to determine which meaning is correct in the given context. This is a key concern for NLP practitioners responsible for the ROI and accuracy of their NLP programs. You can proactively get ahead of NLP problems by improving machine language understanding.

Capturing the information is the easy part but understanding what is being said (and doing this at scale) is a whole different story. Semantic analysis aids in analyzing and understanding customer queries, helping to provide more accurate and efficient support. Continue reading this blog to learn more about semantic analysis and how it can work with examples. With sentiment analysis, companies can gauge user intent, evaluate their experience, and accordingly plan on how to address their problems and execute advertising or marketing campaigns. In short, sentiment analysis can streamline and boost successful business strategies for enterprises. Semantic analysis plays a vital role in the automated handling of customer grievances, managing customer support tickets, and dealing with chats and direct messages via chatbots or call bots, among other tasks.

In the early days of semantic analytics, obtaining a large enough reliable knowledge bases was difficult. In 2006, Strube & Ponzetto demonstrated that Wikipedia could be used in semantic analytic calculations.[2] The usage of a large knowledge base like Wikipedia allows for an increase in both the accuracy and applicability of semantic analytics. For Example, you could analyze the keywords in a bunch of tweets that have been categorized as “negative” and detect which words or topics are mentioned most often. Semantic analysis, on the other hand, is crucial to achieving a high level of accuracy when analyzing text.

Stavrianou et al. [15] present a survey of semantic issues of text mining, which are originated from natural language particularities. This is a good survey focused on a linguistic point of view, rather than focusing only on statistics. The authors discuss a series of questions concerning natural language issues that should be considered when applying the text mining process. Most of the questions are related to text pre-processing and the authors present the impacts of performing or not some pre-processing activities, such as stopwords removal, stemming, word sense disambiguation, and tagging.