Personalized Search Using User Preferences on Social Media

Collaborative Filtering (CF) is a commonly used technique in recommendation systems. It can promote items of interest to a target user from a large selection of available items. It is divided into two broad classes: memory-based algorithms and model-based algorithms. The latter requires some time to build a model but recommends online items quickly, while the former is time-consuming but does not require pre-building time. Considering the shortcomings of the two types of algorithms, we propose a novel Community-based User domain Collaborative Recommendation Algorithm (CUCRA). The idea comes from the fact that recommendations are usually made by users with similar preferences. The first step is to build a user-user social network based on users' preference data. The second step is to find communities with similar user preferences using a community detective algorithm. Finally, items are recommended to users by applying collaborative filtering on communities. Because we recommend items to users in communities instead of to an entire social network, the method has perfect online performance. Applying this method to a collaborative tagging system, experimental results show that the recommendation accuracy of CUCRA is relatively good, and the online time-complexity reduces to O(n).

In social networks, users can easily share information and express their opinions. Given the huge amount of data posted by many users, it is difficult to search for relevant information. In addition to individual posts, it would be useful if we can recommend groups of people with similar interests. Past studies on user preference learning focused on single-modal features such as review contents or demographic information of users. However, such information is usually not easy to obtain in most social media without explicit user feedback. In this paper, we propose a multimodal feature fusion approach to implicit user preference prediction which combines text and image features from user posts for recommending similar users in social media. First, we use the convolutional neural network (CNN) and TextCNN models to extract image and text features, respectively. Then, these features are combined using early and late fusion methods as a representation of user preferences. Lastly, a list of users with the most similar preferences are recommended. The experimental results on real-world Instagram data show that the best performance can be achieved when we apply late fusion of individual classification results for images and texts, with the best average top-k accuracy of 0.491. This validates the effectiveness of utilizing deep learning methods for fusing multimodal features to represent social user preferences. Further investigation is needed to verify the performance in different types of social media.

Recommender systems play an important role in most modern e-commerce applications. They have allowed users to become aware of the myriad choices available to them. The ease of information and the abundance of options have helped users make educated decisions. A recommender system studies a user's preferences and continues learning the user's changing interests, so as to suggest items that incline with the user's interests. In cases where a user is new to the application, or the user prefers not to discourse preferences, the recommender system is unable to gather the user's preference on any item. This is called the cold start problem; wherein the system can make valid recommendations only once the user starts informing the system about his/her choices. In this paper, we discuss the challenges faced by the cold start problem and how this problem may be alleviated using social media. We suggest an approach where we collect public information from users' social media accounts and analyze this information to understand their preferences. In particular, we gather the new user's information using their Twitter profile; i.e., the user's interest and preferences are extracted from his/her Twitter profile by analyzing his/her tweets. These interests will help the system understand what kind of movies the user will be most interested in. We compare these preferences with the metadata about the individual items. Using this approach, we develop a movie recommendation system wherein we produce top-N movie recommendations for a user. We used the MovieTweetings dataset to model the application. Two sets of results have been produced. In the first, smaller set of 770 users, 72.67% of users have received 100% accurate movie recommendations while nearly 80% of users got more than 75% accuracy. For the second, larger set of more than 3,500 users, 53 % of users have received 100% accurate recommendations while 72% of users got more than 75% accuracy. These encouraging results have demonstrated that the approach is effectively in alleviating cold start problems in recommendation systems, and may be applicable to many other e-commerce applications.

The development of big data has brought changes to society and brought us challenges. How to extract useful information from complex data has become the focus of research in recent years. Personalized recommendation as an effective solution has received widespread attention in academia and industry. Collaborative filtering has been widely used by finding users with similar user behaviors as the preferences of similar users. However, the existing methods ignore the interaction information between the user and the item during feature extraction, which leads to imperfect feature extraction and affects the algorithm effect. This paper proposes a graph convolution collaborative filtering model based on the attention model, which uses the graph convolution network to embed user and item interaction information into feature vectors, and uses the attention model to highlight the relatively important interaction information among them, so as to obtain more excellent feature vector. The experimental results show that the model has a good effect on the two commonly used evaluation metrics: recall and normalized discounted cumulative gain(NDCG).

Collaborative filtering (CF), the most successful and widely used technique, recommends items based on the preferences of similar users. The main potentials of CF are its cross-genre recommendation ability, and that it is completely independent of representation of the items being recommended. However, CF suffers from sparsity and cold start problems. On the other hand, a highly effective variant of content-based filtering (CBF), reclusive methods (RMs) based on the preference of the single individual for whom recommendations to be made, provides a methodology that considers uncertainty and the multivalued nature of item features as well as user preferences in a content-based framework using fuzzy logic approaches. The adoption of RM paradigm has several advantages when compared to CF such as sparsity and new item problem, but it suffers from overspecialization and limited content analysis. In view of the complementary nature of CF and RM, we develop a hybrid recommender system (RS) that helps in alleviating aforementioned problems in each approach. First, we propose fuzzy naïve Bayesian classifier based CF (FNB-CF) and RM (FNB-RM) for handling correlation-based similarity problems. To overcome individual weaknesses of FNB-CF and FNB-RM, we develop a hybrid RS, FNB-CF-RM. Effectiveness of our proposed hybrid RS is demonstrated through experimental results using the MovieLens and IMDb data sets.

Current search engine performances need to be improved because often the result suggested by search engine are determine the popularity of a given page for its associated keywords but does not match specific user expectations. Previous researches have indicated that only 20% to 45% of the common search results are relevant. The search becomes harder when the keyword used is a homographic word, and the results have caused user frustration. The work reported in this paper attempt to design Dominant User Context (DUC) Filtering Framework for personalized search in the effort to solve the problem. Literature analysis was performed to identify types of data commonly used for personalized search results. Using the results, this research developed a framework for search result cluster and dominant user contexts, and performed empirical study to validate the framework. The research performed two phases of survey in order to develop search result cluster and test the implementation of DUC. The result provides a novel idea to increase the relevance of search result and contribute to the body of knowledge in Distributed and Parallel Information Retrieval area. The result enables the enhancement of personalized search result by matching the user behavior, interest and ontology of metadata using the search keyword. This contributes to humanizing the search result, in which it will reduce the gap between human and computer. The result also contributes a new conceptual understanding of the types of data commonly used in developing the personalize search result. The developed DUC Filtering Framework provides a foundation in developing the algorithm for searching tools. This algorithm will largely benefit search engine as well as website search function in producing personalized search results.

With the advent of social network services such as Twitter, Facebook, Tumbler, and Google+, the research on social network and media analysis has been greatly advanced. In recent years, the interactions among people, sharing of knowledge and experiences, community activities in social network services increase greatly, which would make the research on social networks more important. Furthermore, as social media contents within social network services are rapidly being produced and consumed, the social media contents now account for the majority of content published on the world wide web. Social media is differentiated from traditional media in many aspects such as its frequency, quality, usability, immediacy, and permanence, which leads to significant potential to the social media analysis research.

Video platforms have become indispensable components within a diverse range of applications, serving various purposes in entertainment, e-learning, corporate training, online documentation, and news provision. As the volume and complexity of video content continue to grow, the need for personalized access features becomes an inevitable requirement to ensure efficient content consumption. To address this need, recommender systems have emerged as helpful tools providing personalized video access. By leveraging past user-specific video consumption data and the preferences of similar users, these systems excel in recommending videos that are highly relevant to individual users. This article presents a comprehensive overview of the current state of video recommender systems (VRS), exploring the algorithms used, their applications, and related aspects. In addition to an in-depth analysis of existing approaches, this review also addresses unresolved research challenges within this domain. These unexplored areas offer exciting opportunities for advancements and innovations, aiming to enhance the accuracy and effectiveness of personalized video recommendations. Overall, this article serves as a valuable resource for researchers, practitioners, and stakeholders in the video domain. It offers insights into cutting-edge algorithms, successful applications, and areas that merit further exploration to advance the field of video recommendation.

Social media is being employed to build support for social, economic, and political justice (Selander et al. 2016; Vaast et al. 2014) in movements such as occupy Wall Street, violence against women's movement and global sustainability movement. These movements have used social media in ways that goes beyond simple communications. As social media allows people to produce and share user generated content, they enable a certain set of affordances of these technologies (Bharati et al 2015; Bharati et al 2014). The social media affordances, available to both collective and individual actors, translate into capabilities afforded to social movements (Tufecki 2014). Research on connective action has examined the effects of digital action repertoires on interaction and engagement such as in the Tea Party and Occupy movements (Agarwal et al. 2014; Selander et al. 2016). Social media can also facilitate mobilization of movement and participation by new volunteers and oftentimes provides a transnational character by diffusing actions beyond the virtual (Van Laer and Van Aelst 2010). Conversation, an essential part of social movements, shapes “social life by altering individual and collective understandings, by creating and transforming social ties, by generating cultural materials that are then available for subsequent social interchange, and by establishing, obliterating, or shifting commitments on the part of participants&x201D; (Tilly 2002, p. 122). In a personal interaction that involves repeated organized interactions between individuals, typically, leads to shared values and trust. The role of social media technologies in furthering this conversation has to be studied and its' influence on social movement ascertained. A few scholars have started to investigate social media affordances and capabilities, especially focused on discourse, during contentious collective action. Still the research has been limited to studying mechanisms of participation, development of a sense of collective identity, creation of community, and framing of political discourse (Farrell 2012; Garrett 2006). Social media data on social movements can involve impersonal “like&x201D; and “share&x201D; to more engaged conversations. Twitter, Instagram, Facebook, and YouTube offer a wide communicative and discourse reach in networks with text, image, audio, and video data. This social media based big data, consisting mostly of unstructured data, comes in the form of social media posts, digital pictures and videos. The symposium will discuss how formal organizational structures and practices might be integrated with social media capabilities to reinforce and enhance social movement organizations as leaders in social change movements. It will explore the characteristics of social media discourse and assess the dynamics of social movements and, subsequent, impact on real-world protests. The symposium will also demonstrate how discourse analysis can be applied visually in order to understand communication patterns. The panel symposium will focus on theoretical and methodological challenges of social media analytics, big data and social movements. Panelists will engage the audience in an interactive discussion on: 1) Theoretical challenges: a. How and why are social movement recruitment and engagement mechanisms being impacted as a result of social media? b. How do we advance theory on social media and social movements when we are overwhelmed with social media based big data? c. What approach should we undertake if big data analysis contradicts most theories on social media and social movements? d. How do we address the issue of generalizability of social media and social movement research when data collection was limited to one social media platform, albeit involving big data? 2) Methodological challenges: a. What methodological approaches have worked in the analysis of social media, big data and social movements? b. How can discourse analysis be applied visually in order to understand communication patterns evident in social media-based big data? c. How can we employ social media analytics to investigate image and video data? d. What combinations of qualitative and quantitative methodologies be employed for big data and social media analytics in the context of social movements? e. What are the limitations of quantitative data analysis techniques, such as structural equation modeling, because of an extremely large sample size? f. What are the limitations of qualitative data analysis techniques as they become extremely labor intensive and, maybe even, impractical because of big data?

This paper proposed a Visual Exploratory Search Engine solution based on cloud computing environment. We focus on improving the traditional search engine by providing a novel framework for people to obtain reliable, personalization and graphical representation search results. The novel strategy is composed of three major parts: 1) Raw information Collect: using novel meta-search engine to quick obtain vast amounts of raw information, 2) Raw information Analysis and Indexing: automatic semantic similarity calculation and semantic link construction based on cloud computing environment, 3) User Interfaces: breakthrough the traditional result use list to represent, we use graphic to display the search result. In comparison with traditional lookup search methodologies, the proposed method is characterized with three main advantages. 1) It infers rich semantic relationships between the query and other related concepts from large-scale meta-search results from Google and Baidu search engines, and representing semantic relationships via graphs, 2) The exploratory search approach enables users to naturally and effectively explore adventure and discover knowledge in a rich information world, 3) And the novel search engine is based on a personalized model, and will provide different user with different search experience. We envision the above method, technology and tools have practical merit in some specific and appropriate application scenario.

Purpose Based on an organizational capability perspective, this paper aims to propose a development model for social media analytics (SMA) capability that can be applied to business-to-business (B2B) marketing, with the aim of facilitating the use and integration of SMA in B2B marketing and maximizing the benefits of business networks in the age of social media. Design/methodology/approach This is a critical interpretive synthesis of SMA publications collected from academic journals, business magazines and the SMA service industry. In addition, an inter-disciplinary approach was adopted by drawing upon both marketing and information systems literature. In total, 123 academic papers, 106 industry case studies and 141 magazine papers were identified and analyzed. The findings were synthesized and compiled to address the predefined research question. Findings An SMA capability development model is proposed. The proposed model consists of four inter-dependent levels (technological, operational, managed and strategic) that collectively transfer the technological capability of SMA to the dynamic organizational capability. Each level of SMA capability is detailed. SMA-in-B2B marketing is highlighted as a socio-technical phenomenon, in which a technological level SMA capability is emphasized as the foundation for developing organizational level SMA capabilities and organizational capabilities, in turn, supporting and managing SMA activities and practices (e.g. strategic planning, social and cultural changes, skills and resources, measurements and values). Practical implications The proposed research framework may have implications for the operational level SMA development and the investigations on the direct and/or indirect measurements to help firms see the impact of SMA on their business. Originality/value This study may have implications for the adoption, use, integration and management of SMA in B2B marketing. The proposed model is grounded on the integrated insights from academia and industry. It is particularly relevant to B2B firms that have engaged in or plan to engage in applying SMA to extract insights from their online networks and is relevant to B2B researchers who are interested in SMA, big data and information technology organization integration studies.

Social media users cannot protect privacy or control the use of internet activity and other personal data collected by social media enterprises. Numerous studies and surveys confirm that social media users and the public at large are increasingly concerned about privacy and activities of social media companies. This author examined what are the societal and user concerns about privacy self-management (PSM), data security, and social media company actions and asks if this suggests that there might be potential market interest in a social media platform to ensure user PSM, control, and monetization their data. The author then reviewed what market-driven applications using existing technology could be used to develop a social media platform to allow the user to ensure privacy and control of PSM and Personal Data, and to monetize activity and static user data. It then developed a conceptual functional model of the platform using existing technology.Social media enterprise business models and technologies drive the revenue strategies of the enterprise. Those models benefit the enterprise’s customers, such as advertisers and merchants, and drive technology design and development to more effectively user data to evaluate, track, and influence those individuals. The users, and their personal data, are the products, not the customers; they have little control or even knowledge about the collection and use of their personal data; and do not realize direct value from its use. These business models create profit for social media enterprises, and increasing concern is being expressed by individuals and governments that commercial excesses do not benefit individuals, create risk, and, through behavior modification, cause actual harm. Yet, the very business models that use free or cheaply acquired personal data to influence individuals make it impossible, or highly difficult, for the industry to become more sensitive to the users’ (in particular, non-paying users) needs, rights, desires, and wellbeing.The author provided a background on the current state of the social media market space. The researcher reviewed published statistics on social media use and privacy, government and legislative actions to regulate social media enterprise activities, and business and technology models that social media enterprises employ to collect and exploit social media user data. The author also provided a background on user preferences and societal concerns about privacy and control of personal data. The researcher conducted a survey of late Millennial and Generation Z (18-39) social media users to gauge market interest in social media services that would safeguard their personal data and enable these users to monetize their data. The author investigated whether there is sufficient interest in an alternative business model supported by existing and new technology that could disrupt the existing social media industry by enabling individuals to gain control, secure, establish ownership of their own user data and create value for themselves by monetizing their own data.The author illustrated a conceptual systems engineering model that integrates existing technology (e.g., open source and licensed software, and commercial services). The researcher examined the feasibility of the model using four methods: conducting a literature search on relevant, existing technologies to protect privacy, control, and monetize social media user data; building a high-level systems model; validating technology choices by review of an expert panel; and validating model functionality by simulating several use cases. The author established that there is almost universal concern about privacy and increasing concern about behavioral manipulation and actual harm caused by social media as well as a desire by users to be able to control their private data. The author concluded that there is a market opportunity for a platform that would enable social media users to control and monetize personal data.The author also determined that there are numerous off-the-shelf open source and commercial technologies available to integrate into a privacy-forward social media platform with a minimum of capital, appropriate for a start-up enterprise. The author concluded by developing a systems model using available technology that a platform that could be built that would enable social media users to control and monetize personal data. <br>

ABSTRACTAssessing the connection between social media engagement and activism for strategic communication is tricky. Although social media facilitate protests, they may also create a lazy form of activism known as slacktivism, potentially weakening the strength of a social movement. This dichotomy suggests the need to develop a theory to understand the connection between social media engagement, empowerment, and protest participation. We explore this connection via in-depth interviews with those who used social media in a case of social media-driven protest. Rather than limit analysis to outcomes like social media analytics, this research suggests the need to consider a model-based understanding of engagement, namely that social media activism proceeds from engagement antecedents and is supported by engagement attributes, towards, ultimately, engagement outcomes. Additionally, data suggest that this process is mediated by factors that include communicative competence, empowerment, and social stake.

Integrating color cues to improve multimodal sentiment analysis in social media

Research on Sentiment Analysis in social media by using Mesopotamian-Iraqi Dialect (MID) of Arabic language was rarely found, there is no reliable dataset developed in MID neither an annotated corpus for the sentiment analysis of social media in this dialect. Therefore, this gap was the main stumbling block for researchers of sentiment analysis in MID, for this reason, this paper introduced the development of an annotated corpus of Mesopotamian-Iraqi Dialect for sentiment analysis in social media and named it as (ACMID) stands for (the annotated corpus of Mesopotamian-Iraqi Dialect) to help researchers in future for using this corpus for their studies, to the best of our knowledge this is the first annotated corpus that both classify polarity as well as emotion classification in MID. Likewise, Facebook as the most popular social platform among Iraqis was used to extract the data from its popular Iraqi pages. 5000 comments were extracted from these pages classified by its polarity (Positive, Negative, Neutral, Spam) by two Iraqi annotators, these annotators were simultaneously classifying the same comments according to Ekman seven universal emotions (Anger, Fear, Disgust, Happiness, Sadness, Surprise, Contempt) or no emotion. Cohen's kappa coefficient was then used to compare the two annotators’ results to find the reliability of these results. The data shows a comparable value among the two annotators for the polarity classification as high as 0.82, while for the emotion classification the result was 0.65.