How different information types affect viewer’s attention on internet advertising

The effect of internet advertising has been a controversial issue, especially on the topic of how to effectively draw more attention from internet users. According to traditional attention theory, we know people pay lesser attention on other objects if the main browsing contents occupy more of the viewer’s mental resources. Therefore, we know different information types of webpage should have different influences on users’ attention. On the other hand, an effect called ‘banner blindness’ makes viewers naturally overlook the advertising based on their previous experience. It is therefore becoming more and more difficult to increase viewer’s attention on advertisement simply by adding salient features on the advertisements. In light of this new challenge in Internet advertising, verifying the different influences of the information types on advertising attention is the main goal of this study.Great amounts of previous studies relevant to internet advertising focused on the advertisement itself, like the form, color, size and location. However, this study put focus on how the information types and the webpage structure influence the viewer’s attention on banner advertising. This research tested the effect on user attention of four common information types on Internet webpages: (1) text-based webpage; (2) text-picture mixed webpage; (3) picture-based webpage; and (4) video-based webpage. This study hopes to provide valuable information for matching advertising with viewing tasks that will stimulate the most user attention.     

Object-based visual query suggestion

State-of-the-art visual search systems allow to retrieve efficiently small rigid objects in very large datasets. They are usually based on the query-by-window paradigm: a user selects any image region containing an object of interest and the system returns a ranked list of images that are likely to contain other instances of the query object. User’s perception of these tools is however affected by the fact that many submitted queries actually return nothing or only junk results (complex non-rigid objects, higher-level visual concepts, etc.). In this paper, we address the problem of suggesting only the object’s queries that actually contain relevant matches in the dataset. This requires to first discover accurate object’s clusters in the dataset (as an offline process); and then to select the most relevant objects according to user’s intent (as an on-line process). We therefore introduce a new object’s instances clustering framework based on a major contribution: a bipartite shared-neighbours clustering algorithm that is used to gather object’s seeds discovered by matching adaptive and weighted sampling. Shared nearest neighbours methods were not studied beforehand in the case of bipartite graphs and never used in the context of object discovery. Experiments show that this new method outperforms state-of-the-art object mining and retrieval results on the Oxford Building dataset. We finally describe two object-based visual query suggestion scenarios using the proposed framework and show examples of suggested object queries.     

基于用户评价的查询串与搜索结果特征权重计算

提出了利用大量用户评价结果来进行特征权重的计算方法,用于解决搜索引擎中查询串与搜索结果的相似度分析。该方法完全利用用户对搜索结果的“潜在评价”来进行。用户对输入查询串所做的点击反映了其内部的关联性,该文提出的方法可获取这种关联性,对该问题建立了数学模型,利用EM算法解决了特征权重的计算。由于模型的函数比较复杂,难于计算其收敛性,因此,使用了模拟退火算法作为EM算法的补充,用于验证算法的收敛性。实验使用百度搜索引擎在竞价广告上进行,提取的测试数据样本为100个广告和144 132个query,获得的数据结果显示,所有特征收敛到全局最优解,抽样部分数据获得检索相似准确率为93.32%,召回率为87.43%。     

Search personalization through query and page topical analysis

Thousands of users issue keyword queries to the Web search engines to find information on a number of topics. Since the users may have diverse backgrounds and may have different expectations for a given query, some search engines try to personalize their results to better match the overall interests of an individual user. This task involves two great challenges. First the search engines need to be able to effectively identify the user interests and build a profile for every individual user. Second, once such a profile is available, the search engines need to rank the results in a way that matches the interests of a given user. In this article, we present our work towards a personalized Web search engine and we discuss how we addressed each of these challenges. Since users are typically not willing to provide information on their personal preferences, for the first challenge, we attempt to determine such preferences by examining the click history of each user. In particular, we leverage a topical ontology for estimating a user’s topic preferences based on her past searches, i.e. previously issued queries and pages visited for those queries. We then explore the semantic similarity between the user’s current query and the query-matching pages, in order to identify the user’s current topic preference. For the second challenge, we have developed a ranking function that uses the learned past and current topic preferences in order to rank the search results to better match the preferences of a given user. Our experimental evaluation on the Google query-stream of human subjects over a period of 1 month shows that user preferences can be learned accurately through the use of our topical ontology and that our ranking function which takes into account the learned user preferences yields significant improvements in the quality of the search results.     

Privacy leakage of certificateless public key authenticated searchable encryption via frequency analysis: Attacks and revises

Certificateless public key authenticated searchable encryption (CLPASE) is a versatile asymmetric searchable encryption that enables ciphertext retrieval, resists inside keyword guessing attacks, and avoids both certificate management problem and key escrow problem. However, most existing CLPASE schemes are vulnerable to frequency analysis which can extract keywords from user-generated trapdoors (i.e., search queries) and thus compromise user’s search privacy.In this paper, we give a detailed analysis showing that most CLPASE schemes reveal the underlying frequency distribution of the target keywords in the trapdoors searched by users, regardless of whether the trapdoor generation algorithm is deterministic or not. The analysis shows that frequency analysis has become a significant threat to users’ search privacy in the CLPASE system. To address this issue, we provide a concrete CLPASE scheme against frequency analysis. We then compare our scheme with previous CLPASE schemes in terms of features and performance evaluation. As a result, our scheme provides higher guarantee for user’s search privacy with comparable efficiency.     

Assisting web search using query suggestion based on word similarity measure and query modification patterns

One of the useful tools offered by existing web search engines is query suggestion (QS), which assists users in formulating keyword queries by suggesting keywords that are unfamiliar to users, offering alternative queries that deviate from the original ones, and even correcting spelling errors. The design goal of QS is to enrich the web search experience of users and avoid the frustrating process of choosing controlled keywords to specify their special information needs, which releases their burden on creating web queries. Unfortunately, the algorithms or design methodologies of the QS module developed by Google, the most popular web search engine these days, is not made publicly available, which means that they cannot be duplicated by software developers to build the tool for specifically-design software systems for enterprise search, desktop search, or vertical search, to name a few. Keyword suggested by Yahoo! and Bing, another two well-known web search engines, however, are mostly popular currently-searched words, which might not meet the specific information needs of the users. These problems can be solved by WebQS, our proposed web QS approach, which provides the same mechanism offered by Google, Yahoo!, and Bing to support users in formulating keyword queries that improve the precision and recall of search results. WebQS relies on frequency of occurrence, keyword similarity measures, and modification patterns of queries in user query logs, which capture information on millions of searches conducted by millions of users, to suggest useful queries/query keywords during the user query construction process and achieve the design goal of QS. Experimental results show that WebQS performs as well as Yahoo! and Bing in terms of effectiveness and efficiency and is comparable to Google in terms of query suggestion time.     

Efficient Monte Carlo clustering in subspaces

One of the key difficulties for users in information retrieval is to formulate appropriate queries to submit to the search engine. In this paper, we propose an approach to enrich the user’s queries by additional context. We used the Language Model to build the query context, which is composed of the most similar queries to the query to expand and their top-ranked documents. Then, we applied a query expansion approach based on the query context and the Latent Semantic Analyses method. Using a web test collection, we tested our approach on short and long queries. We varied the number of recommended queries and the number of expansion terms to specify the appropriate parameters for the proposed approach. Experimental results show that the proposed approach improves the effectiveness of the information retrieval system by 19.23 % for short queries and 52.94 % for long queries according to the retrieval results using the original users’ queries.     

Unintrusive customization techniques for Web advertising

Most online advertisement systems in place today use the concept of consumer targeting: each user is identified and, according to his or her system setup, browsing habits and available off-line information, categorized in order to customize the advertisements for highest user responsiveness. This constant monitoring of a user's online habits, together with the trend to centralize this data and link it with other databases, continuously nurtures fears about the growing lack of privacy in a networked society. In this paper, we propose a novel technique of adapting online advertisement to a user's short term interests in a non-intrusive way. As a proof-of-concept we implemented a dynamic advertisement selection system able to deliver customized advertisements to users of an online search service or Web directory. No user-specific data elements are collected or stored at any time. Initial experiments indicate that the system is able to improve the average click-through rate substantially compared to random selection methods.     

Index Navigator: Search Engine with Reasoning for Understanding and Expressing User's Changing Mind

When a search engine user becomes interested in a new area for him/herself, it is difficult for the user to enter a query precisely expressing the interest or to select areas including the interest, because he/she is just a beginner of the interest. This paper presents a system called Index Navigator, which tells areas a user is interested in, keywords he/she should enter as a query, and documents concerning his/her interest. A tough problem for such a system is to understand the user's interest from the query he/she entered. Index Navigator employs an inference method called Cost-based Cooperation of Multi-Abducers (CCMA), for understanding a user's interest from the history of the user's queries (expression of interest in incomplete keywords), even if the changing speed of the user's interest can not be estimated. With this device, Index Navigator guided the user to areas, keywords and documents relevant to his/her interest, according to the experimental results.     

Attention for Web Directory Advertisements: A Top-Down or Bottom-Up Process?

Web directories have attracted many advertisers with their special advantages in their large user base. Until recently, attention mechanism of advertisements (ad) on web directories is not well understood. To investigate how the ad location and color of web directories influence users’ attention, this study uses eye tracking to measure the participants’ search time, total fixation duration and the location of the first fixation. Results reveal that visual attention on the ad area of a web directory is user-driven and follows a top-down process. The location of users’ first-fixations is the center of the screen. Ad links that place in the center area and on the top-left corner would increase users’ notice. Ad links that change color in the center area have the advantages of attracting user attention. Our findings suggest that ad links should be placed in the center area or on the top-left corner to increase users’ notice. Ad links placed in the center area should be designed using salient color to catch users’ visual attention.     

不确定性区域上的多维关键词搜索

随着在线地图应用的普及,基于地图的空间对象检索成为一个重要的工具而被广泛使用,技术也比较成熟。人们在地图上经常进行确定性目标点查询,例如用户提交关键词“咖啡店”,地图应用会在地图上标记所有的咖啡店,用户还可以通过进一步操作获取咖啡店的详细信息。但实际生活中存在另一种需求,例如用户想找到一个区域,在这个区域内要有“咖啡店”、“学校”和“旅店”这三类对象,称这样的查询为不确定性区域检索查询。目前对地图应用的研究无法解决不确定性区域检索的问题。而利用矩形剪枝和top-k推荐能够通过用户提交的关键字,给用户返回若干候选区域。     

L-Shape advertising for mobile video streaming services: Less intrusive while still effective

Advertising is significant in video streaming services on mobile devices, as well as in other e-business services. Advertisement providers pay part or the entire service cost; hence, users enjoy services at a low price or free of charge. However, users perceive advertisements as clutter and avoid them. To reduce advertising avoidance and enhance advertising efficiency, various methods, such as raising advertising relevance, seeking advertisement permission, or changing advertisement position, are suggested. This study introduces L-Shape advertising, a new strategy suitable for mobile video streaming services, as it is less distractive and effective. We demonstrate the superiority of L-Shape advertising compared to popular advertising formats such as skippable and non-skippable advertisements. The study empirically tests how the proposed strategy performs and examines users’ cognitive processes applying eye-tracking technology. We found that L-Shape advertising is less distractive but still effective.     

Searching for people to follow in social networks

With the development of social networks, more and more users have a great need to search for people to follow (SPTF) to receive their tweets. According to our experiments, approximately 50% of social networks’ lost users leave due to a lack of people to follow. In this paper, we define the problem of SPTF and propose an approach to give users tags and then deliver a ranked list of valuable accounts for them to follow. In the proposed approach, we first seek accounts related to keywords via expanding and predicting tags for users. Second, we propose two algorithms to rank relevant accounts: the first mines the forwarded relationship, and the second incorporates the following relationship into PageRank. Accordingly, we have built a search system¹ that to date, has received more than 1.7 million queries from 0.2 million users. To evaluate the proposed approach, we created a crowd-sourcing organization and crawled 0.25 billion profiles, 15 billion messages and 20 billion links representing following relationships on Sina Microblog. The empirical study validates the effectiveness of our algorithms for expanding and predicting tags compared to the baseline. From query logs, we discover that hot queries include keywords related to academics, occupations and companies. Experiments on those queries show that PageRank-like algorithms perform best for occupation-related queries, forward-relationship-like algorithms work best for academic-related queries and domain-related headcount algorithms work best for company-related queries.     

基于联合概率矩阵分解的上下文广告推荐算法

上下文广告与用户兴趣及网页内容相匹配,可增强用户体验并提高广告点击率.而广告收益与广告点击率直接相关,准确预测广告点击率是提高上下文广告收益的关键.目前,上下文广告推荐面临如下问题:(1) 网页数量及用户数量规模很大;(2) 历史广告点击数据十分稀疏,导致点击率预测准确率低.针对上述问题,提出一种基于联合概率矩阵分解的因子模型AdRec,它结合用户、广告和网页三者信息进行广告推荐,以解决数据稀疏时点击率预测准确率低的问题.算法复杂度随着观测数据数量的增加呈线性增长,因此可应用于大规模数据.     

Efficient fuzzy full-text type-ahead search

Traditional information systems return answers after a user submits a complete query. Users often feel “left in the dark” when they have limited knowledge about the underlying data and have to use a try-and-see approach for finding information. A recent trend of supporting autocomplete in these systems is a first step toward solving this problem. In this paper, we study a new information-access paradigm, called “type-ahead search” in which the system searches the underlying data “on the fly” as the user types in query keywords. It extends autocomplete interfaces by allowing keywords to appear at different places in the underlying data. This framework allows users to explore data as they type, even in the presence of minor errors. We study research challenges in this framework for large amounts of data. Since each keystroke of the user could invoke a query on the backend, we need efficient algorithms to process each query within milliseconds. We develop various incremental-search algorithms for both single-keyword queries and multi-keyword queries, using previously computed and cached results in order to achieve a high interactive speed. We develop novel techniques to support fuzzy search by allowing mismatches between query keywords and answers. We have deployed several real prototypes using these techniques. One of them has been deployed to support type-ahead search on the UC Irvine people directory, which has been used regularly and well received by users due to its friendly interface and high efficiency.     

空间关键字个性化语义近似查询方法

现有的空间关键字查询处理模式大都仅支持位置相近和文本相似匹配,但不能将语义相近但形式上不匹配的对象提供给用户;并且,当前的空间-文本索引结构也不能对空间对象中的数值属性进行处理。针对上述问题,本文提出了一种支持语义近似查询的空间关键字查询方法。首先,利用词嵌入技术对用户原始查询进行扩展,生成一系列与原始查询关键字语义相关的查询关键字;然后,提出了一种能够同时支持文本和语义匹配,并利用Skyline方法对数值属性进行处理的混合索引结构AIR-Tree;最后,利用AIR-Tree进行查询匹配,返回top-k个与查询条件最为相关的有序空间对象。实验分析和结果表明,与现有同类方法相比,本文方法具有较高的执行效率和较好的用户满意度;基于AIR-Tree索引的查询效率较IRS-Tree索引提高了3.6%,在查询结果准确率上较IR-Tree和IRS-Tree索引分别提高了10.14%和16.15%。     

Density-based spatial keyword querying

With the rocket development of the Internet, WWW(World Wide Web), mobile computing and GPS (Global Positioning System) services, location-based services like Web GIS (Geographical Information System) portals are becoming more and more popular. Spatial keyword queries over GIS spatial data receive much more attention from both academic and industry communities than ever before. In general, a spatial keyword query containing spatial location information and keywords is to locate a set of spatial objects that satisfy the location condition and keyword query semantics. Researchers have proposed many solutions to various spatial keyword queries such as top-K keyword query, reversed kNN keyword query, moving object keyword query, collective keyword query, etc. In this paper, we propose a density-based spatial keyword query which is to locate a set of spatial objects that not only satisfies the query’s textual and distance condition, but also has a high density in their area. We use the collective keyword query semantics to find in a dense area, a group of spatial objects whose keywords collectively match the query keywords. To efficiently process the density based spatial keyword query, we use an IR-tree index as the base data structure to index spatial objects and their text contents and define a cost function over the IR-tree indexing nodes to approximately compute the density information of areas. We design a heuristic algorithm that can efficiently prune the region according to both the distance and region density in processing a query over the IR-tree index. Experimental results on datasets show that our method achieves desired results with high performance.