Please enable JavaScript.
Coggle requires JavaScript to display documents.
Sentiment Analysis Over Social Networks: An Overview (Information…
Sentiment Analysis Over Social Networks: An Overview
Information
The rapid increase in data on social media creates a need for mining such data to get valuable insights.
The data type can be unstructured with large volumes. Sentiment analysis addresses such need by detecting opinions or emotions on the social media text. Sentiment analysis can be performed in various domains such as social, medical and industrial applications.
Sentiment analysis can be applied in four levels: sentence, aspect and document and user level. This can be performed using machine learning (clustering or classification), lexicon, NLP, Ontology or hybrid techniques.
There are many enhancement methods to enhance sentiment analysis results such as feature selection, data integration, data cleaning, and crowdsourcing.
Feature selection
Used for choosing suitable features from
text that enhance sentiment analysis results
Techniques
Crowdsourcing
Public crowd experience
Data cleaning
Data integration
Spam or fake sentiment detection in reviews or posts is an important application of the the sentiment analysis[11], [8]. Sentiment analysis also can be used to define trust over social network for a brand or a service [12] or to build recommendation systems [13], [4], which recommend a service, a place or a product for user.
Sentiment analysis addresses such need by detecting opinions or emotions on the social media text
Sentiment Analysis Level
Level 1 / sentence level
Level 2 / document level
Level 3 / aspect level
it is used in case of the availability of attributes inside entity, post or input text. Each attribute can hold a sentiment in its own. For example, a customer review on a mobile phone has the attributes battery life, screen light and other attributes. Each attribute can have a different sentiment
The aspect level can lead to a better analysis and results if taken into consideration.
Example : My phone is really nice but I have a bad battery. It contains slow applications but I am happy with its screen.
The aspect here is the phone while the attributes are battery, applications, and screen. Sentiment detection can lead to the following results (battery, negative), (application, negative), (screen, positive).
Some sentiment analysis techniques apply grouping on the aspect level where all attributes having the same sentiment result are grouped together. The grouping of the previous example will lead to the following result: (battery, application, and negative) and (screen, Positive).
Level 4 / user level
Handles the social relationships between different users using graph theory
Example: A is a user who has a friend B connected to him. User B is always mentioned in user A posts, always gains likes and shares from user A.. User A might have the same opinion or sentiment as user B. This can be the result of the influence of user B on user A and how much such user can affect user B opinion. The user level takes such influence into consideration.
SENTIMENT ANALYSIS METHODS
There are two types of lexicons [17]. The first type is corpus lexicon, which is divided into two types (semantic oriented, statistical oriented). Corpus lexicon, such as SenticNet [12] can achieve more accurate sentiment results as it is context oriented not similarity of words oriented. An example of semantic oriented lexicons can be found in [18] where the authors dealt with the meaning of words based on a concept net lexicon. On the other hand, in [19] the authors presented a statistical method in defining sentiments.
Sentiement analysis lexicons
The second type of lexicons is dictionary based. In [20], two dictionaries were presented. The first is a word dictionary, which ripped with human emotion. The second is a topic modeling or a topic oriented dictionary, which is helpful in aspect sentiment analysis. Some researchers [21] tackles dictionary based lexicons by integrating existing seeds or dictionaries to build more valuable multi domain dictionaries.
SENTIMENT ANALYSIS ENHANCEMENT METHODS
Sentiment anaylsis data cleaning
Data cleaning [31], [8] is an important preprocessing phase which enhances sentiment analysis results. Data cleaning operations include tokenizing, stemming and filtering. Data cleaning can be applied in two phases [31], data transformation and data filtering.
Data transformation [31], [8] operations involve but are not limited to removing useless spaces, handling abbreviations and negations, stemming and removing stop words. Data filtering [31] is related to selecting features which are suitable for sentiment analysis.
Dimension Reduction
Process of reducing high dimensions using two methods either feature selection or feature extraction.
Feature extraction
Transformative method which applies a transformation on the data to project it into a new feature space with lower dimensions.
Feature selection
The process of selecting features from the original data set based on specific selection criteria taking into consideration that the result subset has the smallest classification error with lossless content meaningure.
Algorithms
Chi-square,
Latent semantic indexing
Point-wise Mutual Information (PMI).
Sentiment Analysis and Crowdsourcing
Crowdsourcing is the science of resolving a problem or task by the help of crowd [5]. Crowdsourcing can help in providing more accurate sentiment analysis results.
Crowd can help in assigning labels to training data set or giving feedback about sentiment classification results, which can enhance the predication and the classification models.
In [7] authors proved that the use of crowdsourcing resulted in more accurate sentiment detection for topic and sentiment classification in social media data.
Sentiment Analysis and Ontologies
Sentiment Analysis and Data Intergation.
Data is integrated, from different sentiment lexicons for sentiment analysis classification. This integration is performed by combining, filtering and deleting the duplicated data from individual dictionaries
Using data-driven or data integration lexicons builds a high quality sentiment lexicon, which enhances the sentiment detection results
Sentiment Analysis And Spam Detection
Characteristics to define fake sentiment include but are not limited to speed of publishing a tweet/post , the tweet/post location fake writing identity, number of mentions, number of hash tags, emotions, URLs in tweet/post and number of tweets/posts or posts per day
Fake and spam sentiments on social media leads to inaccurate
sentiment detection results
In [36], [11] the authors extracted geographic user characteristics and tweets contentbased features to discover spam sentiment within text
Building user profiles [36] over social networks or defining online identity can help in detecting fake spam sentiment contents and fake spam users.
Sentiment Analysis And User Profiling
In [37], the authors proved that there is a strong relationship between online user identity and the users contribution in blogs.
Online user identity and user profiling helps in making the sentiment analysis results more accurate as it measures polarity based on the user profile in addition to the post polarity
They divided the online users into classes based on their online features like (kindness, social skills, creativity). Others presented profiling models [37], [38] to either predict the political interests of users or data publishing interests [38].
Sentiment Analysis And Text Summarization
A hybrid system for target or aspect oriented summarization was presented in [38] where the authors defined features of the aspect (product or service) and applied sentiment analysis to classify the available sentiments.
Sentiment summarization is the process of summarizing sentiment according to a specific domain or a topic, also called target based summarization
Another attempt of summarization was carried out in [17] where a summarization of Arabic tweets was performed to generate specific topics rather than reading all the tweets.
Meta
Author
Khaled Ahmed
Neamat El Tazi
Ahmad Hany Hossny
Year
2015
Goals
Presents a survey about sentiment analysis addressing the different concepts in this area, problems and its solutions, available APIs, tools used and presenting a list of open challenges in this area.
Result