Many websites encourage people to submit reviews of various products and services. We present and evaluate a novel approach to efficiently model and analyze the text within user reviews to estimate how much reviewers care about different aspects of a product (i.e., amenities, food, location, room, etc. of a hotel). Our approach performs statistically quite similar to the best existing method. However, our method for computing aspect weights is a linear time method while the current state of the art solution requires cubic time at best.

We study the problem of automatically extracting information networks formed by recognizable entities as well as relations among them from social media sites. Our approach consists of using state-of-the-art natural language processing tools to identify entities and extract sentences that relate such entities, followed by using text-clustering algorithms to identify the relations within the information network. We propose a new term-weighting scheme that significantly improves on the state-of-the-art in the task of relation extraction, both when used in conjunction with the standard tf ċ idf scheme and also when used as a pruning filter. We describe an effective method for identifying benchmarks for open information extraction that relies on a curated online database that is comparable to the hand-crafted evaluation datasets in the literature. From this benchmark, we derive a much larger dataset which mimics realistic conditions for the task of open information extraction. We report on extensive experiments on both datasets, which not only shed light on the accuracy levels achieved by state-of-the-art open information extraction tools, but also on how to tune such tools for better results.

Financial transaction systems, such as those utilizing the Financial Information Exchange (FIX) protocol, must make buy/sell decisions based on information received over a network interface. However, at the time that a decision to buy or sell is made, some data on the system may not yet be visible to the financial engine software because it has not yet completed being processing by the network protocol stack. This data is hidden from a FIX engine running in user space until the network protocol processing (in the kernel) is fully completed and the message contents are copied from kernel space to user space. This hidden data may contain information that would lead to a more advantageous transaction if it were available to the financial application at the time that the trade decision is made. We have made several modifications to the TCP/IP protocol stack in the Linux operating system. Our protocol modifications allow a user application to determine if there are pertinent pending messages currently being processed in the kernel. If this is the case, the user application may then choose to delay the buy/sell decision by a short time until these additional messages have completed kernel processing and become available to the user application. We have implemented this mechanism under Linux 2.6.27.24 and are evaluating alternate implementations and the significance of the decision improvements that result from this enhancement.

We describe our experiences in applying learning-to-rank techniques to improving the quality of search results of an online hotel reservation system. The search result quality factors we use are average booking position and distribution of margin in top-ranked results. (We expect that total revenue will increase with these factors.) Our application of the SVMRank technique improves booking position by up to 25% and margin distribution by up to 14%.

Approximately 74% of queries in peer-to-peer file sharing systems return no results, wasting both network bandwidth and computational resources. Considering the popularity of this application, this is a significant problem. We propose to improve the accuracy of query routing by improving how file collections are described and defining how peers organize into groups. Our basic approach is to improve the precision of collection description by having each peer partition its collection and describe each collection independently. Experimental results on real peer-to-peer file sharing data shows that our techniques can reduce routing cost to peer groups by 40% with near-perfect load balance.

The state-of-the-art in Named Entity Recognition relies on a combination of local features of the text and global knowledge to determine the types of the recognized entities. This is problematic in some cases, resulting in entities being classified as belonging to the wrong type. We show that using global information about the corpus improves the accuracy of type identification. We explore the notion of a global domain frequency that relates relation identifying terms with pairs of entity types which are used in that relation. We use this to identify entities whose types are not compatible with the terms they co-occur in the text. Our results on a large corpus of social media content allows the identification of mistyped entities with 70% accuracy.

An increasing amount of Web information is in video format. Today's search technology allows videos to be found using graphical features and textual descriptions. However, the information gleaned from video features is coarse, while textual descriptions are often short and fail to capture the precise content of videos. We hypothesize that user comments contain supplemental information that effectively describes the content of a video. This information, once extracted, can be applied to a search engine index to improve video search accuracy.

Abstract Peer‐to‐peer (P2P) file sharing is a leading Internet application. Millions of users use P2P file‐sharing systems daily to search for and download files, accounting for a large portion of Internet traffic. Due to their scale, it is important to fully understand how these systems work. We analyze user queries and shared files collected on the Gnutella system, draw some conclusions on the nature of the application, and propose some research problems.

Spam is highly pervasive in P2P file-sharing systems and is difficult to detect automatically before actually downloading a file due to the insufficient and biased description of a file returned to a client as a query result. To alleviate this problem, we first characterize spam and spammers in the P2P file-sharing environment and then describe feature-based techniques for automatically detecting spam in P2P query result sets. Experimental results show that the proposed techniques successfully decrease the amount of spam by 9 % in the top-200 results and by 92 % in the top-20 results.

Structured peer-to-peer networks support keyword search by building a distributed index over the collective content shared by all peers. Building the index and processing queries involve data transfer among peers, thus it is important to keep both of these activities bandwidth-efficient. However, this goal is difficult to attain, as smaller, less precise indices reduce index building and access costs but increase query processing cost, which potentially increases overall cost. We study the trade-off between indexing cost and query processing cost in a structured peer-to-peer network and propose a cost-reducing, adaptive, distributed indexing technique based on the term distributions in local shared contents and user query logs. Using this information, we reduce costs by tuning the precision of the index. The approach we take is to group local documents and to index the groups instead of either individual documents or entire peer collections. We control total cost by controlling the number and contents of groups. We propose a probabilistic model to estimate the cost of grouping, which allows us to identify the optimal number of groups to be created. In addition, we propose a cost-based distance function to guide the document grouping process. Experimental results show that our adaptive indexing technique reduces cost by up to 47% compared with peer-level grouping and by up to 73% compared with document-level grouping.

Query routing in peer-to-peer systems is based onthe peers' content summaries. For the sake of scalability, summaries are built at a "peer level." The coarseness of peer level summaries limits their ability to route queries accurately to peers sharing relevant contents. We propose to improve the quality of each peer's summary by grouping its contents and summarizing each group individually. To enhance grouping effectiveness, we incorporate information about real-world term distributions into group generation. Experimental results show that our techniques significantly improve query routing accuracy with little cost.

Clustering large data sets with high dimensionality is a challenging data-mining task. This paper presents a framework to perform such a task efficiently. It is based on the notion of data space reduction, which finds high density areas, or dense cells, in the given feature space. The dense cells store summarized information of the data. A designated partitioning or hierarchical clustering algorithm can be used as the second step to find clusters based on the data summaries. Using Kmeans as an example, this paper presents GARDEN-Kmeans, which performs data space reduction using Gamma Region DENsity partition, and utilizes Kmeans to cluster the summarized information. The experimental study shows that GARDEN-Kmeans executes several orders of magnitude faster than basic Kmeans and the recursive bisection Kmeans algorithm of CLUTO, while producing comparable clustering quality

We introduce IR-Wire, a tool for information retrieval research and education in peer-to-peer file-sharing systems. Built on top of Lime Wire's implementation of the popular Gnutella standard, it includes functionality to collect data on queries and shared files and stores them in a way to make analyses simple. IR-Wire is designed modularly to facilitate its customization for other uses.

Peer-to-peer file-sharing systems suffer from the over-specification of query results due to the fact that queries are conjunctive and the descriptions of shared files are sparse. Ultimately, longer queries, which should yield more accurate results, actually do the opposite. The judicious masking of query terms circumvents the shortcomings of conjunctive query processing, significantly improving query accuracy.

Participants in a decentralized system often use some local ranking information in selecting effective collaborations. We say that such systems are preference-based for most practical types of preferences, such systems converge towards a unique stable configuration. In this paper, we investigate the speed and quality of the convergence process with respect to the model parameters. Our results provide insight into the design of system parameters, such as the number of connections or the algorithm for choosing new partners.

Many studies focus on the Web, but yet, few focus on peer-to-peer file-sharing system queries despite their massive scale in terms of Internet traffic. We analyzed several million queries collected on the Gnutella network and differentiated our findings from those of Web queries.

One of the reasons that search performance of peer-to-peer file-sharing systems is poor is that results are poorly described, hurting the client's ability to rank them. This is a consequence of the independent and distributed nature of servers in this architecture. We propose automatically probing the network for better descriptions to improve result ranking. Search performance improvements are demonstrated through extensive simulation results

Peers in peer-to-peer file-sharing systems use conjunctive queries as a way of controlling query cost in the absence of information about the behavior of other peers. Conjunctive queries are good at increasing the precision of query result sets, but may be overly selective, decreasing overall performance. We consider relaxing the conjunctive matching criterion and its impact on performance and cost. Experimental results indicate that significant performance improvements are possible at reasonable cost

This paper introduces a new algorithm for clustering data in high-dimensional feature spaces, called GARDENHD. The algorithm is organized around the notion of data space reduction, i.e. the process of detecting dense areas (dense cells) in the space. It performs effective and efficient elimination of empty areas that characterize typical high-dimensional spaces and an efficient adjacency-connected agglomeration of dense cells into larger clusters. It produces a compact representation that can effectively capture the essence of data. GARDENHD is a hybrid of cell-based and density-based clustering. However, unlike typical clustering methods in its class, it applies a recursive partition of sparse regions in the space using a new space-partitioning strategy. The properties of this partitioning strategy greatly facilitate data space reduction. The experiments on synthetic and real data sets reveal that GARDENHD and its data space reduction are effective, efficient, and scalable.

Synchronization performance is a major problem with intermittently connected mobile databases. A server periodically generates update files for each client, which are downloaded and applied when convenient. Unfortunately, the time required to synchronize clients in this way increases drastically with client population. We show that this trend could be altered by appropriately modifying the way that update files are designed, resulting in significant performance improvements.

Peer-to-peer file-sharing systems have hundreds of thousands of users sharing petabytes of data, however, their search functionality is limited. In general, query results contain many references to the same data object. These references are grouped, and the size of the group - the number of references it contains - metric. Although group size is effective in finding popular data, it works poorly for rare, less popular data. Other ranking functions, such as precision and cosine similarity, are more appropriate in this case. The authors showed the significant performance benefit in finding rare data using these ranking functions through extensive simulation.

This paper studies fast access to data that are broadcast on multiple channels. Broadcast is a useful data dissemination technique because of its scalability, but is lacking when it comes to response time. Increasing the number of available broadcast channels is a logical way of increasing throughput. Little work, however, has considered the access structures necessary for making effective use of the additional channels. We propose various indexing schemes for a multi-channel broadcast program. We demonstrate the effectiveness of our techniques in decreasing response time and tuning time via extensive experiments over a wide range of parameters.

We give an overview on techniques for quickly finding items that are broadcast on multiple data channels. Previous works assume that throughput increases with the number of data channels because the total available bandwidth increases. This assumption, however, rests on the client's ability read data as soon as it is broadcast on any channel. Without this ability, the benefit of additional channels is compromised as the search space for data also increases. We describe the consequences of increasing the number of broadcast channels on search. We then offer candidate search techniques and show their effects on three metrics: search time, tuning time, and hop count.

We introduce the linear file template, which is a generic data organization suitable for use with many different types of data. The linear file is specifically designed to handle intense database update loads concurrently with processing of analytic queries.

Broadcast is a scalable way of disseminating data because broadcasting an item satisfies all outstanding client requests for it. However, because the transmission medium is shared, individual requests may have high response times. In this paper, we show how to minimize the average response time given multiple broadcast channels by optimally partitioning data among them. We also offer an approximation algorithm that is less complex than the optimal and show that its performance is near-optimal for a wide range of parameters. Finally, we briefly discuss the extensibility of our work with two simple, yet seldom researched extensions, namely, handling varying sized items and generating single channel schedules.

To avoid the high cost of continuous connectivity, a class of mobile applications employs replicas of shared data that are periodically updated. Updates to these replicas are typically performed on a client-by-client basis--that is, the server individually computes and transmits updates to each client--limiting scalability. By basing updates on replica groups (instead of clients), however, update generation complexity is no longer bound by client population size. Clients then download updates of pertinent groups. Proper group design reduces redundancies in server processing, disk usage and bandwidth usage, and dimininishes the tie between the complexity of updating replicas and the size of the client population. In this paper, we expand on previous work done on group design, include a detailed I/O cost model for update generation, and propose a heuristic-based greedy algorithm for group computation. Experimental results with an adapted commercial replication system demonstrate a significant increase in overall scalability over the client-centric approach.

Article Free Access Share on A framework for designing update objects to improve server scalability in intermittently synchronized databases Authors: Wai Gen Yee College of Computing, Georgia Institute of Technology, 801 Atlantic Drive, Atlanta, GA College of Computing, Georgia Institute of Technology, 801 Atlantic Drive, Atlanta, GAView Profile , Michael J. Donahoo Baylor University, P.O. Box 97356, Waco, TX Baylor University, P.O. Box 97356, Waco, TXView Profile , Shamkant B. Navathe College of Computing, Georgia Institute of Technology, 801 Atlantic Drive, Atlanta, GA College of Computing, Georgia Institute of Technology, 801 Atlantic Drive, Atlanta, GAView Profile Authors Info & Claims CIKM '00: Proceedings of the ninth international conference on Information and knowledge managementNovember 2000 Pages 54–61https://doi.org/10.1145/354756.354787Published:06 November 2000Publication History 4citation388DownloadsMetricsTotal Citations4Total Downloads388Last 12 Months16Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteeReaderPDF