Add Thesis

Web 2.0: Improving customer experience with wedding service providers through investigation of the ranking mechanism and sentiment analysis of user feedback on Instagram

Written by M. Jäderlund

Paper category

Bachelor Thesis


Computer Science




Instagram is one of the main social platforms for commercial promotion [1]. Millions of potential customers and endless visual marketing opportunities make Instagram an ideal place to increase online sales. There are many tools and mechanisms to promote a brand on Instagram, such as paid advertising or using a set of pre-generated popular tags. In this regard, the existence and content of user reviews have become important social and psychological factors for the motivation to purchase or use products or services. The goal of this degree program is to study natural language processing techniques applied to Instagram user comments to determine a new algorithm that incorporates content analysis into the list of feed ranking factors. Just like now, users must read through the posts on Instagram to understand the quality of the product or service. Therefore, a method for classifying and ranking products and services is needed. We have proposed a new algorithm called "Wed 2.0" that can help consumers search for wedding services and products on Instagram. Data mining technology [2] and sentiment analysis [3] are used to define the sentiment of comments and build user opinions, and to rank accounts based on this knowledge. Instagram [5] As one of the most popular social networks, it pays close attention to search functions and content discovery. Instagram search results are divided into four categories: top, person, hashtag, and location (see Figure 1.1). The top tab displays popular accounts, hashtags, and geotags, which contain some or all of the words from the search query. The people tab can be conveniently used to search for people by nickname. The Tags tab displays all publications with topic tags that contain words from the search query. The location label shows the location that contains the word from the search query. The standard Instagram search bar algorithm presents search results based on a variety of factors, including Instagram and Facebook activities, user interests, engagement, and the number of followers and likes in the account [6]. Due to the various services of buying fans and likes on Instagram, the latter two may be misleading. The goal of the project is to explore another recommendation method based on user feedback in the comments section. The idea is to present search results based on customer satisfaction with a particular product rather than based on the number of fans and likes. The more positive feedback an account receives, the higher its ranking in search results. The project focuses on data extraction and data mining technology, natural language processing and sentiment analysis. Extracting data from the Web or "Web scraping" [7] is a method of retrieving data from a Web source. It allows us to manually or automatically extract new or updated data and save it for later use. There are many Read Less