PDF Publication Title:
Text from PDF Page: 001
MEASURING INFLUENCE ON INSTAGRAM: A NETWORK-OBLIVIOUS APPROACH 1 Measuring Influence on Instagram: a Network-oblivious Approach Noam Segev, Noam Avigdor, Eytan Avigdor Abstract—This paper focuses on the problem of scoring and ranking influential users of Instagram, a visual content sharing online social network (OSN). Instagram is the second largest OSN in the world with 700 million active Instagram accounts, 32% of all worldwide Internet users. Among the millions of users, photos shared by more influential users are viewed by more users than posts shared by less influential counterparts. This raises the question of how to identify those influential Instagram users. In our work, we present and discuss the lack of relevant tools and insufficient metrics for influence measurement, focusing on a network oblivious approach and show that the graph-based approach used in other OSNs is a poor fit for Instagram. In our study, we consider user statistics, some of which are more intuitive than others, and several regression models to measure users’ influence. Index Terms—Social Media, Instagram, Influence, Ranking 3 1 INTRODUCTION T H E he transition to Web 2.0 transformed the business models of online marketing from a global ad approach based to individual opinions and targeted campaigns [1], [2], [3], [4]. Web 2.0 not only took traditional marketing strategies to the extreme via viral marketing campaigns [5], [6], [7], but it also gave rise to new techniques of brand building and audience targeting via influencer marketing [8], [9]. In fact, the use of micro-influencers, trusted indi- viduals within their communities, has been seen as a more effective way to build a brand in terms of audience reception and return on investment [10], [11], [12]. Instagram, which is a visual content sharing online social network (OSN), has become a focal point for influencer marketing. With power users and micro-influencers pub- lishing sponsored content companies need to rate these influencers and determine their value [13], [14], [15]. Most of today’s scoring themes rely on graph-based algorithms of a known network graph. Such graphs are not always available, and building them for Instagram users requires a great deal of resources, e.g., crawling time and computing costs. A possible solution would be to infer the underlying network structure using the user activity logs, as described by Barbieri et al. [16], but even in the event a graph is constructed it would not necessarily be of much use given that information decays exponentially along the graph even under optimal passive information propagation, which is not the case. The rest of the paper is organized as follows: In Section 2 we described OSNs in greater detail as well as current influ- ence measuring schemes. We then present our annotations and formal description of the problem of measuring and ranking influence in Section 3. The dataset of Instagram users and their posts is described in Section 4, followed by discussion on the extracted and aggregated features of the testable data in Section 4.2. Following this, we present our testing methodology, baselines, regression models and • N. Segev, N. Avigdor and E. Avigdor are with Klear Influencer Marketing E-mail: [noam.segev, n, e]@klear.com experimental results in Section 6. Finally, we discuss our conclusion and possible future work in Section 7. 2 BACKGROUND AND RELATED WORK Online social media networks are often described as a di- rected graph with entities such as users acting as nodes and relationships as the edges. Such edges can be unidirectional or bi-directional, e.g., an Instagram ”follower” and a Face- book ”friendship”, respectively. These edges do not need to represent a long-lasting relationship; they can signal a one- time engagement, e.g., a ”like” or a ”comment”. Following this, link prediction in OSNs became an active research field focused on community detection, in the case of users as nodes [17], [18], or content suggestion otherwise [19], [20], [21]. In most OSNs, user-generated content is “pushed”, i.e., propagated via interaction. When a user uploads a post, their followers can see the post and choose to pass it along, creating a pyramid-formed cascade of information. Thus, if user A follows user B who, in turn, follows user C, and user C posts some content user B chooses to share, user A is passively influenced by user C. These social micro-networks tend to grow around influential, active users [22], [23]. Instagram content, however, is “pulled”, i.e., information propagation requires activity along the pyramid, such that, using our earlier example, for user C’s post to reach user A, user A must look for content suggested by trusted users. This situation raises the question of how to rank users in OSNs. As OSNs are traditionally described as graphs, ranking has been done using various graph statistics, from simple in/out degree to node closeness [24], [25], as is the case with the work of Anger and Kittl in Twitter [26] and Agarwal et al. in the context of influential blogs [27]. Other techniques extend to existing link analysis algorithms - the most popular one being PageRank [28], [29]. Weng et al. suggested twitterRank [30] and Khrabrov et al. introduced starRank [31], both extensions of PageRank working on Twitter’s follower and engagements graphs, respectively. arXiv:1806.00881v1 [cs.SI] 3 Jun 2018PDF Image | MEASURING INFLUENCE ON INSTAGRAM: A NETWORK-OBLIVIOUS APPROACH
PDF Search Title:
MEASURING INFLUENCE ON INSTAGRAM: A NETWORK-OBLIVIOUS APPROACHOriginal File Name Searched:
instagram-measuring-influence.pdfDIY PDF Search: Google It | Yahoo | Bing
Cruise Ship Reviews | Luxury Resort | Jet | Yacht | and Travel Tech More Info
Cruising Review Topics and Articles More Info
Software based on Filemaker for the travel industry More Info
The Burgenstock Resort: Reviews on CruisingReview website... More Info
Resort Reviews: World Class resorts... More Info
The Riffelalp Resort: Reviews on CruisingReview website... More Info
| CONTACT TEL: 608-238-6001 Email: greg@cruisingreview.com | RSS | AMP |