Abstract: In at least one embodiment, a log-structured system for metadata (MD) updates can be leveraged to enable use of variable size MD pages with multiple different encoding formats. Processing can include: generating an updated MD page by performing one or more updates to an existing version of the MD page; selecting a first page type to be used for encoding the updated MD page, wherein said selecting selects the first page type from a plurality of predefined page types, wherein each of the plurality of predefined page types identifies a different metadata layout or format for storing MD pages; encoding the updated MD page in accordance with the first page type and accordingly generating a first encoded version of the updated MD page; and storing the first encoded version of the updated MD page at a first physical storage location of a non-volatile storage device.

Abstract: In a compaction scheme for a log structured merge (LSM) tree, a storage device is configured to: receive a first user data piece from a host; generate a first meta data piece for a highest level of the LSM tree, corresponding to the first user data piece; when the highest level exceeds a set storage limit, trigger a compaction process on the first and second meta data pieces to generate compacted meta data pieces excluding overlapping meta data elements of the second meta data pieces overlapped with the first meta data piece. Through a garbage collection, victim user data elements corresponding to the overlapping meta data elements are deleted.

Abstract: Data set discovery is disclosed, including: identifying first file metadata elements for a first file associated with a node in a hierarchy of data; identifying second file metadata elements for a second file associated with the node; identifying common file metadata elements among the first file metadata elements and the second file metadata elements; and determining that the common file metadata elements represent a data set comprising at least the first file and the second file.

Abstract: A computing system can be configured to input model input that includes context data into a machine-learned model and receive model output that describes one or more semantic entities referenced by the context data. The computing system can be configured to provide data descriptive of the semantic entity or entities to the computer application(s) and receive application output(s) respectively from the computing application(s) in response to providing the data descriptive of semantic entity or entities to the computer application(s). The application output(s) received from each computer application can describe available action(s) of the corresponding computer application with respect to the semantic entity or entities. The computing system can be configured to provide at least one indicator to a user that describes the available action(s) of the corresponding computer applications with respect to the semantic entity or entities.

Abstract: Techniques are described herein that are capable of performing an aggregation-based determination of resiliency of a cloud computing service to chaos events. Chaos events are executed against a cloud computing service during a time frame that includes multiple time periods. For each chaos event, service health indexes (SHIs) that correspond to the time periods are calculated. Each SHI indicates an extent to which indicators, which represent respective measurements of performance of the cloud computing service during the respective time period, satisfy respective criteria. The SHIs for each chaos event are analyzed to provide a respective resiliency score. At least some of the resiliency scores of the respective chaos events are aggregated to provide an aggregated resiliency score, which represents resiliency of the cloud computing service to the chaos events.

Abstract: Exemplary methods and systems provide for production of geographically located, accurate indoor maps of physical spaces and provide navigation guidance for users in physical spaces, and process requests for users related to available physical space within said physical spaces. Aggregated information about the physical spaces is curated and converted using automated procedures into a data model that contains all elements of said physical spaces, including geographic data about the physical space's location on the globe. The system and computer program embodiments retrieve the converted data, load it on mobile devices or computers, rendering it on these devices in the correct global geographic context, allowing users to interact with the data to select rooms, and send and receive requests to and from external systems regarding these selections. Users navigate within the physical spaces using mobile devices and the computer program embodiments in tandem to provide accurate directions and guidance.

Abstract: A method and apparatus for indicating the status of an ancillary embedded system in an electronic device. In one exemplary embodiment, the method includes starting an initialization process of a high-level embedded system in the electronic device. The method further includes determining the status of the ancillary embedded system. The method further includes generating display information for the status of the ancillary embedded system. The method further includes storing the display information in a manner retrievable by the high-level embedded system. The method further includes reading the stored set of display information and displaying an indication of the status on a user display prior to completion of the high-level embedded system's initialization process. The method further includes periodically updating the stored set of display information by the ancillary embedded system to provide a real-time indication of status.

Abstract: Content management systems that utilize external content libraries are described herein. According to some embodiments, an example system includes a content management system configured to create web content that includes at least one asset that resides on an external system that is remote to the system, the web content having a reference to the at least one asset, an external content library connector system exposing connectors that are accessible to both the content management system and a dynamic experience delivery system, and the dynamic experience delivery system configured to invoke the external content library connector system to obtain the at least one asset based on the reference using one or more of the connectors and publish the web content with the at least one asset included therein.

Abstract: Metadata of a root node from a serialized representation of a file system metadata tree data structure is obtained. A representation of the root node that references a stub intermediate node that is a placeholder for an intermediate node stored in the serialized representation is created. A request to access data of a leaf node in a tree path including the intermediate node is received. In response to the request, metadata of the intermediate node is obtained from serialized metadata representation.

Abstract: Features are disclosed for facilitating remote management of network directories of organizations by a directory management system. The network directories may change over time, experiencing growth in size and number of current connections, increased latency, reduced performance, and the like. The network directories may also shrink over time, experience fewer connections, etc. Organizations can define scaling policies by which the directory management system can automatically respond to the occurrence of various events, such as changes in the size or usage of the organizations' network directories, by scaling resources associated with the directories. The directory management system can perform various scaling actions on-demand or without requiring additional action by the organizations, thereby reducing the time and effort required by the organizations to monitor their own directories and implement (or request implementation of) changes.

Abstract: In non-limiting examples of the present disclosure, systems, methods and devices for surfacing and interacting with electronic meeting insights are presented. A meeting request that includes a plurality of invitees may be received. The meeting request may be associated with a plurality of related documents based on one or more attributes and a meeting insight related to the related documents may be surfaced. The meeting request may have a low acceptance in relation to a threshold and a corresponding insight may be surfaced to reschedule the meeting. A user may send a time modification suggestion to the meeting organizer and a rescheduling insight may be surfaced. Time and location information corresponding to invitees and the meeting time may be analyzed and reminders may be surfaced for each of the invitees in relation to a meeting location. Attendees may interact with location-based insights to have directions to a meeting surfaced.

Abstract: Techniques are described herein that are capable of performing an aggregation-based determination of resiliency of a cloud computing service to chaos events. Chaos events are executed against a cloud computing service during a time frame that includes multiple time periods. For each chaos event, service health indexes (SHIs) that correspond to the time periods are calculated. Each SHI indicates an extent to which indicators, which represent respective measurements of performance of the cloud computing service during the respective time period, satisfy respective criteria. The SHIs for each chaos event are analyzed to provide a respective resiliency score. At least some of the resiliency scores of the respective chaos events are aggregated to provide an aggregated resiliency score, which represents resiliency of the cloud computing service to the chaos events.

Abstract: A method for storing mobile device data to cloud storage includes detecting a predetermined condition to backup data associated with a first mobile device to the cloud storage. Uploading, to a cloud storage server, data associated with the first mobile device including data associated with a plurality of third-party applications stored on the first mobile device and one or more of contacts information stored on the first mobile device, text messages stored on the first mobile device, or contents of a memory card associated with the first mobile device. The data associated with the plurality of third-party applications include binary data and metadata. The method further includes storing the binary data of the plurality of third-party applications at a first cloud storage location and storing the metadata of the plurality of third-party applications to a second cloud storage location that is different from the first cloud storage location.

Abstract: Provided is an information processing system, and the like, with which even if target information that relates to a target of interest has been disguised, such disguising can be more accurately detected, and the appropriate information intended for presentation to a user can be presented according to the detection results. An information processing system is equipped with: a server apparatus which determines whether or not target information is authentic, on the basis of a determination condition and information from which target information relating to a target of interest has been created and which indicates at least a time/date and a first location and a client which controls a presentation unit such that presentation information intended for presentation to a user is presented according to the result of the determination made by the server apparatus.

Abstract: Storing an entity as an object or a file in a hybrid storage system based on an accessing protocol is presented herein. In response to determining that a request to modify a data entity is based on an object storage protocol, and in response to determining that the data entity is a first object that is represented as a first file in a file storage device, the hybrid storage system creates the first object in an object storage device and deletes the first file. Further, in response to determining that the request is based on a file storage protocol, and that the data entity is a second file that is represented as a second object in the object storage device, the hybrid storage system creates the second file in the file storage device and deletes the second object.

Abstract: Described herein is a system and method of connectivity migration of an executing virtual application and/or guest operating system. State associated with a first instance of an application and/or a guest operating system executing on a first virtual machine is captured. Information regarding connectivity state associated with a plurality of running connections between the first virtual machine and client device(s) is also captured (e.g., layers 2, 3 and 4). The captured state information can be provided to a second virtual machine which utilizes the captured station information to establish state for a second instance of the application, a second instance of the guest operating system, and/or connectivity of the plurality of running connections between the second virtual machine and client device(s). The state of the second instance of the application can be synchronized with the state of the second instance of the guest operating system.

Abstract: A computer-implemented method according to one embodiment includes determining a starting folder within a file system, computing, for each child folder of the starting folder, a similarity metric indicating a level of similarity to a file, selecting two child folders of the starting folder having greatest similarity metrics, comparing a difference between the greatest similarity metrics of the two child folders to a predetermined threshold, and conditionally selecting the starting folder as a recommended folder to which the file is saved, based on the comparing.

Abstract: Systems and methods are disclosed for a unified file index for a file system. In one example, a Grand Unified File Index (GUFI) includes a tree replicating the directory hierarchy of one or more primary filesystems, and individual metadata stores for each directory. The GUFI tree permits fast traversal, efficient user space access controls, and simple tree directed operations such as renames, moves, or permission changes. In some examples, the individual metadata stores can be implemented as embedded databases on flash storage for speed. In some examples, use of summary tables at the directory or subtree level can eliminate wasteful executions, prune tree traversal, and further improve performance. In various examples, efficient operation can be achieved from laptop to supercomputer scale, across a wide mix of file distributions and filesystems.

Abstract: Managing metadata added to a file, a directory, or a symbolic link is provided. One or more pieces of metadata of a movement target are obtained. The movement target is at least one of a file, a directory, and a symbolic link. The movement target is to be moved from one file system to another file system within a plurality of file systems. Whether the one or more pieces of metadata are migratable to the other file system is determined based, at least in part, on metadata stored in the other file system. The one or more pieces of metadata are processed by (i) migrating a first piece of metadata that is determined to be migratable to the other file system and (ii) leaving a second piece of metadata in the one file system, wherein the second piece of metadata is determined to be non-migratable.

Abstract: According to aspects illustrated herein, methods and systems for retaining duplex scanning information in a scanned output file, is disclosed. The method includes receiving a document for scanning, having multiple pages, wherein each page has a first side such as a front side and a second side such as a back side. Then, a check for selection of a duplex scanning feature by a user, is performed. Upon scanning, an image data representing an image of the document is generated. The image data and the duplex scanning feature selected by the user is processed to generate a scanned output file. The duplex scanning information is generated, while generating the scanned output file. The duplex scanning information is retained in the scanned output file, for later retrieval and use. The duplex scanning information helps identify the first side and the second side in the scanned output file.

Abstract: An apparatus in one embodiment comprises a multi-tier storage system having at least a front-end storage tier, a back-end storage tier and a data mover module configured to control movement of data between the storage tiers. The multi-tier storage system is configured to receive a plurality of files, such as relatively small size files containing data from respective Internet of Things (IoT) devices. The front-end storage tier comprises a file system and at least one key-value store. The multi-tier storage system is configured to combine multiple ones of the received files written to at least one specified directory of the file system into an object for storage in association with a key in the key-value store. The object comprises multiple sub-keys identifying respective ones of the multiple files within the object. The data mover module may be configured to copy the object from the key-value store to the directory.

Abstract: Systems and methods for generating a template directory based on an existing directory using the hierarchy of files in the directory. The system may use clustering analysis to determine similarities between items within the directory or files to generate a profile based on the characteristics of the saved items. The template directory may then be moved to another computing environment and items on the new computing environment may be arranged in a structure the mimics the existing directory. The system may update the profile of files in the directory dynamically as new items are created and saved within the computing environment hosting the template directory.

Abstract: Disclosed in the embodiments of the disclosure are methods and apparatuses for grouping web page labels. The method comprises: extracting titles of a plurality of web page labels in a tab bar of a browser; calculating semantic distances between the extracted plurality of titles; clustering web page labels corresponding to the plurality of titles based on the semantic distances between the plurality of titles; obtaining at least one label group based on the clustering, the one label group comprising at least one web page label; and sequentially arranging the at least one label group in the tab bar, the web page labels belonging to the same label group being successively arranged in the tab bar. After grouping, the web page labels are still presented in the format of web page labels in the tab bar and a web page label can be selected with a single click.

Abstract: In the field of government engagement management, for users of an employee desktop web client, it is now possible, within the web client application, to search and read articles and/or knowledge content that has been authored to external locations. Due to this integration to external, third-party applications, content and/or articles can be displayed to an agent on the employee desktop web client graphical user interface. Agents can enter free text into a specific search field and review the results in summary form, and then select an article in HTML format to progress the current interaction with the client. An additional feature extending from this capability is to add an amount of coding to external knowledge content websites that are owned and/or operated by the owner of the system such that when the website is viewed through the third-party integration module, a button or icon appears within the website that when selected takes the agent to an appropriate form.

Abstract: A mobile terminal is capable of automatically editing icons corresponding to a plurality of applications included in a home screen page. The mobile terminal includes: a touch screen configured to output a home screen page including icons corresponding to applications; and a controller configured to generate a folder including at least part of the applications by using attribute information of the applications, when an edition request with respect to the icons included in the home screen page is received, wherein an image of the generated folder, rather than the icons corresponding to the applications included in the generated folder, is output to the home screen page.

Abstract: The disclosed computer-implemented method for evaluating security services may include (i) receiving, at a backend security server from an enterprise, multiple suspicious computing events detected within the enterprise, (ii) recording, within the backend security server, historical security information for each computing event that includes (a) a classification of the computing event as malicious or non-malicious based on a security analysis performed by the backend security server and (b) a point in time at which the classification was determined, (iii) evaluating an ability of the backend security server to detect security threats by (a) detecting an additional computing event within the enterprise and (b) determining, based on the historical security information, a point in time at which the backend security server became capable of classifying the additional computing event, and (iv) adjusting a security policy within the enterprise based on the evaluated ability of the backend security server.

Abstract: A system and method for pruning data based on metadata. The method may include receiving a query that includes a plurality of predicates and identifying one or more applicable files including database data satisfying at least one of the plurality of predicates. The identifying the one or more applicable files including reading metadata stored in a metadata store that is separate from the database data. The method further includes pruning inapplicable files comprising database data that does not satisfy at least one of the plurality of predicates to create a reduced set of files and reading the reduced set of files to execute the query.

Abstract: A system and method for efficient cache flushing are provided. The disclosed method includes maintaining a data structure in connection with a plurality of blocks used for data caching, the data structure including a row lock wait list section. The method further includes receiving an Input/Output (I/O) request, performing a hash search for the I/O request against the data structure, and based on the results of the hash search, locking at least one row in a data cache thereby preventing read and write operations from being performed on the at least one row until the at least one row is unlocked.

Abstract: In particular embodiments, a consent receipt management system is configured to track and analyze one or more attributes of a user interface via which data subjects are requested to provide consent (e.g., consent to process, collect, and/or store personal data) in order to determine which of the one or more attributes are more likely to result in a successful receipt of consent from a data subject. For example, the system may be configured to analyze one or more instances in which one or more data subjects provided or did not provide consent in order to identify particular attributes and/or factors that may increase a likelihood of a data subject providing consent.

Abstract: A method for creating a common platform graphical user interface is provided. The interface may enable a user to trigger a data load job from a tool. The tool may monitor file upload events, trigger jobs and identify lists of missing or problematic file names. The tool may run on a single thread, thereby consuming relatively less system resources than a multi-thread program to perform its capabilities. The tool may enable selection of file names using wildcard variables or keyword variables. The tool may validate a list of files received against a master file list for each data load job. The tool may receive user input relating to each data load job. The tool may generate a loop within the single thread to receive information. The tool may analyze the received information and use the received information to predict future metadata associated with future data load jobs.

Abstract: Aspects of the disclosure relate to live reporting of check image keying issues. A computing platform may receive a first dataset and may transmit a first set of instructions for triggering transaction balancing and the first dataset to a first client computer system. Subsequently, the computing platform may receive from the first client computer system, a request to report an issue. Responsive to receiving the request to report an issue, the computing platform may receive from the first client computer a second dataset relating to the issue. Subsequently, the computing platform may generate, based on the first dataset and second dataset, an issue record, and may store the issue record in a second database. Then, the computing platform may transmit, via the communication interface, a second set of instructions for triggering issue investigation and the issue record relating to the issue to a second client computer system.

Abstract: Features are disclosed for facilitating remote management of network directories of organizations by a directory management system. The network directories may change over time, experiencing growth in size and number of current connections, increased latency, reduced performance, and the like. The network directories may also shrink over time, experience fewer connections, etc. Organizations can define scaling policies by which the directory management system can automatically respond to the occurrence of various events, such as changes in the size or usage of the organizations' network directories, by scaling resources associated with the directories. The directory management system can perform various scaling actions on-demand or without requiring additional action by the organizations, thereby reducing the time and effort required by the organizations to monitor their own directories and implement (or request implementation of) changes.

Abstract: In embodiments of statistics time chart interface cell mode drill down, a first interface displays in a table format that includes columns each having a column heading comprising a different value, each different value associated with a particular event field, and includes one or more rows, each row having a time increment and aggregated metrics that each represent a number of events having a field-value pair that matches the different value represented in one of the columns and within the time increment over which the aggregated metric is calculated. A cell can be emphasized that includes one of the aggregated metrics in a row that includes the respective time increment, and in response, a menu displays options to transition to a second interface.

Abstract: A framework (referred to herein as Application Integration Architecture, or AIA) that formalizes and orchestrates activities in an SOA development lifecycle. In one set of embodiments, AIA can capture development-related information in a shared data store and cause the information to flow in an automated or semi-automated manner from one lifecycle phase to the next as the lifecycle progresses. This information flow can, in turn, facilitate automations at each lifecycle phase for the responsible stakeholders (e.g., solution architects, developers, installation developers, etc.), thereby enforcing SOA best practices, enhancing development productivity, and ensuring the quality of the final SOA deliverables.

Abstract: A computer-implemented method operable in a content delivery service (CDN), includes, by a collector system: receiving multiple event streams of event data, said multiple event streams comprising event data from a plurality of CD services in said CDN, each event of said event streams comprising: (i) a timestamp for said event, (ii) information relating to said event; and producing state data relating to information represented in said event data of said multiple event streams while being able to asynchronously respond to queries relating to said state data.

Abstract: In an infrastructure that uses a mobile order fulfillment system, robotic drive units may be dispatched and instructed to bring inventory holders to a workstation where at least one of the inventory holders is packed and prepared for shipment. The robotic drive units are then instructed to move the prepared inventory holder to a transport vehicle such as a truck. Fiducial marks may be removably placed within the transport vehicle to aid navigation of the robotic drive units. At a destination facility, additional robotic drive units may be instructed to move the inventory holders from the truck and place the inventory holders at appropriate storage locations.

Abstract: Various embodiments of the present invention provide a mobile computing device that operates multiple, co-existing and independent operating system environments on a common kernel. A booting process for initiating a multiple operating system environment is also provided. Additionally, various embodiments of the present invention include processes for managing a switch between one operating system environment to a second operating system environment.

Abstract: A device with multiple, co-existing, and independent environments interacting with a common kernel, and related methods of operation, are disclosed herein. Operation is altered or dependent on the device being or entering a docked mode.

Abstract: Various embodiments of the present invention provide a mobile computing device that operates multiple, co-existing and independent operating system environments on a common kernel. A booting process for initiating a multiple operating system environment is also provided. Additionally, various embodiments of the present invention include processes for managing a switch between one operating system environment to a second operating system environment.

Abstract: Technology for serving events to a plurality of client computing devices is disclosed. In some embodiments, the technology implements a queue containing a set of elements comprising one or more NULL components; and a list of events, each event corresponding to one or more client updates received from the associated client computing devices. The technology associates pointers in the queue with client computing devices for processing updates, and uses the pointers to determine whether or not to send particular updates to particular clients, thereby preventing some clients from slowing down update processing for all clients.

Abstract: A method and software is described for recreating on a target datastore a set of hierarchical files that are present on a source datastore. A content identifier (ID) is maintained for each component of the set of hierarchical files. The content ID of a component is updated when its contents are modified. The child component is copied from the source datastore to the target datastore. The content ID corresponding to the parent component on the source datastore is compared with content IDs corresponding to files present on the target datastore. When a matching content ID is discovered, it infers a copy of the parent component. The matching file on the target datastore is associated with the copied child component so that the matching file becomes a new parent component to the copied child component, thereby recreating the set of hierarchical files on the target.

Abstract: One example method presents a synthetic view of data stored by a shared device file system. The synthetic view provides a “fictional” single namespace for data stored by the shared device file system. The method includes accepting a command to access data stored by the shared device file system. The command may be native to a user requesting the access. The command addresses the data stored by the shared device file system using the single namespace provided by the synthetic view. The method includes creating a replacement command(s) for accessing data stored by the shared device file system using one or more actual namespaces different from the single fictional namespace provided by the synthetic view. With the replacement command(s) prepared, the method proceeds to access data stored by the shared device file system using the replacement command(s).

Abstract: Methods, systems, and user interfaces for accessing, retrieving, storing, and presenting information stored in computer readable media. Information stored on a computer readable media may be searched using selectable items representing an index of information. In response to selection of one of the selectable items, information is presented which is related to the selected item, concurrently with display of the index and at least one least one previously-selected item in a hierarchical display. The invention has particular application to the legal field, but is also applicable to other fields, e.g., accounting, medicine, business and finance, engineering, scientific fields, merchandising fields, fulfillment fields, inventory fields, etc.

Abstract: Disclosed are various embodiments for storage of files. A removable memory device is configured to couple to a computing device, and a storage management application is stored in the removable memory device, the storage management application being executable by a processor circuit. The storage management application is configured to send a plurality of files for storage in a networked storage system, the networked storage system being remote to the computing device. The storage management system caches a subset of the files on the removable memory device and maintains a local file directory in the removable memory device. The local file directory lists the files stored in the networked storage system in association with an account linked to the removable memory device.

Abstract: There is provided a method implemented in a data storage system including a plurality of storage nodes connected to each other via a communications network. At least one storage node of the plurality of storage nodes may store a first data item. The data item may be arranged to include a reference to a second data item stored in the data storage system. The method may include providing an identifier of the second data item. The method may also include providing status information about the first data item. The identifier may be used to update the first data item, for example based on the status information. The first data item may be stored at the at least one storage node.

Abstract: A non-transitory computer-readable recording medium has stored therein a generating program that causes a computer to execute a process including: selecting a control mode from a first control mode and a second control mode in accordance with whether a document element that includes a predetermined number or more of child elements is present in a document file, the first control mode and the second control mode determining storing locations of data in the document file from a plurality of blocks, the first control mode determining the storing locations of data in the document file for each document element in a hierarchy of the child elements, and the second control mode determining the storing locations of data in the document file for each document element in the hierarchy of the document element that includes the child elements or in the hierarchy higher than the hierarchy of the document element.

Abstract: Storing and retrieving user data files includes, in a file storing operation, applying a first mapping (e.g. hash function) to a file identifier to originate a content address in a flat content address space; applying a second mapping (e.g., grouping of digits/characters) to the content address to originate a multi-level directory name usable in a file system; invoking the file system to store the user data file in the named directory. A file retrieving operation includes applying the first mapping to the file identifier to regenerate the content address; applying the second mapping to the regenerated content address to regenerate the directory name; invoking the file system using the regenerated directory name to retrieve the user data file. The method enables a general-purpose network attached storage (NAS) system to store files organized according to a legacy content-addressed storage (CAS) model of existing applications.

Abstract: According to one embodiment, an electronic device has an interface, storage and a controller. The interface is configured to connect with an external storage device. The storage is configured to store data. The controller is configured to determine whether information for identifying a folder name is stored in the external storage device when the external storage device is connected to the interface, and to recognize the external storage device as a folder having the folder name corresponding to the information when the information is stored in the external storage device.

Abstract: A method of aggregating and using medical device data from a plurality of remote institutions. The system and method electronically receives at a central computer system a plurality of established medical device data, each of the plurality of established medical device data being received from a respective medication delivery system, each of the respective medication delivery systems having a respective plurality of medical devices within the respective remote institution, such as medication delivery pumps, associated therewith and utilized therein. The system and method electronically combines and stores the plurality of established medical device data from each of the plurality of remote institutions within a memory, and electronically provides a remote client computer access to at least one of a central reporting application adapted for providing summary information to the remote client computer about the medical device data, and/or other applications.

Abstract: This invention discloses an image retrieval apparatus. The image retrieval apparatus comprises an unlabelled image selector for selecting one or more unlabelled image(s) from an image database; and a main learner for training in each feedback round of the image retrieval, estimating relevance of images in the image database and a user's intention, and determining retrieval results, wherein the main learner makes use of the unlabelled image(s) selected by the unlabelled image selector in the estimation. In addition, the image retrieval apparatus may also include an active selector for selecting, in each feedback round and according to estimation results of the main learner, one or more unlabelled image(s) from the image database for the user to label.