Abstract: A method includes obtaining, at a device, a hierarchical structure representing a graphical layout of content items of an electronic document, the content items including at least text. The method also includes generating a word embedding representing a word of the electronic document. The method further includes determining position information of a location of the word in the electronic document. The method also includes determining a descriptor that indicates a relationship of the location to the hierarchical structure. The method further includes providing input data to a machine learning model to generate a semantic region category label of a semantic region of the electronic document. The semantic region includes the word. The input data includes the word embedding, the position information, and the descriptor.

Abstract: A method of generating a searchable representation of an electronic document includes obtaining an electronic document specifying a graphical layout of content items. The content items include at least text in a table. The method also includes selecting masking rules, generating a vertical mask based on the masking rules, and generating a horizontal mask based on the masking rules. The vertical mask indicates estimated locations of vertical boundaries of table columns of the table, and the horizontal mask indicates estimated locations of horizontal boundaries of table rows of the table. The method also includes identifying cells of the table based on the vertical mask and the horizontal mask and generating a searchable data structure based on text corresponding to the identified cells of the table.

Abstract: A method includes receiving input designating a term of interest in a document of a document corpus and determining a target context embedding representing a target word group that includes the term of interest and context words located in the document proximate to the term of interest. The method also includes identifying, from among the document corpus, a first candidate word group that is semantically similar to the target word group and a second candidate word group that is semantically dissimilar to the target word group. The method further includes receiving user input identifying at least a portion of the first candidate word group as associated with a first label and identifying at least a portion of the second candidate word group as not associated with the first label. The method also includes generating labeled training data based on the user input to train a text classifier.

Abstract: A method for measuring sea state includes scanning an area of a waterborne moving object over a period of time using at least one sensor to obtain point cloud data of the moving object. The area of the moving object is identified based on the point cloud data. Changes in movement of the area of the moving object are characterized over the period of time based on the point cloud data to calculate a state of the moving object. Sea state is estimated based on the state of the moving object.

Abstract: A method of detecting a landing zone includes scanning an area using a sensor system to obtain data of the area. One or more markings are identified from the data. The one or more markings are verified as corresponding to an intended landing zone.

Abstract: A method for measuring sea state includes scanning an area of a waterborne moving object over a period of time using at least one sensor to obtain point cloud data of the moving object. The area of the moving object is identified based on the point cloud data. Changes in movement of the area of the moving object are characterized over the period of time based on the point cloud data to calculate a state of the moving object. Sea state is estimated based on the state of the moving object.

Abstract: A method of detecting a landing zone includes scanning an area using a sensor system to obtain data of the area. One or more markings are identified from the data. The one or more markings are verified as corresponding to an intended landing zone.

Abstract: A noise producing fan with low air movement or variable air movement is provided. The device comprises a conventional mechanical fan that is adapted to provide a soothing and steady volume of ambient noise for a user to block out background noise and sleep comfortably. The mechanical fan includes several embodiments, including different fan blade arrangements, means of interchanging fan blades on the mechanical fan hub, different fan shrouds that allow for access to the fan hub, and finally an imbalanced fan hub that creates an oscillatory motion to increase noise production. The fan blades for low air movement and efficient noise generation are flat or perforated, and secure to the fan hub as it oscillates and rotates within the mechanical fan housing.

Abstract: A fastener includes a head and a threaded shank. The shank defines an inner periphery and the thread defines an outer periphery. The shank and the thread each have a generally polygonal crossection with at least five lobes separated by at least five convex surfaces.

Abstract: A method and apparatus for producing randomly variegated multiple strand wrapped yarn in twisting together two or more yarns at a plurality of yarn twisting stations. At each station a pneumatic twisting head is disposed in which randomly turbulent air currents are created in a chamber in the twisting head by compressed air flowing into the chamber from a manifold through bores in the twisting head, thereby randomly twisting together two or more yarns in the chamber.

Abstract: A method and apparatus for producing randomly variegated multiple strand wrapped yarn in twisting together two or more yarns at a plurality of yarn twisting stations. At each station a pneumatic twisting head is disposed in which randomly turbulent air currents are created in a chamber in the twisting head by compressed air flowing into the chamber from a manifold through bores in the twisting head, thereby randomly twisting together two or more yarns in the chamber.

Abstract: A method and apparatus for producing randomly variegated multiple strand yarn in twisting together two or more yarns at a plurality of yarn twisting stations. At each station a pneumatic twisting head is disposed in which randomly turbulent air currents are created in a chamber in the twisting head by compressed air flowing into the chamber from a manifold through bores in the twisting head, thereby randomly twisting together two or more yarns in the chamber. The yarns which have been twisted together are taken up over a plurality of traversing drums commonly mounted on a shaft which is driven by an alternating electric current motor controlled by a control device which operates to start and restart the motor on a predetermined cycle including a variation of the electrical input sufficient to result in randomly unpredictable inertia resistance of the traversing drums to rotational speed variation, thereby creating randomly unpredictable take-up of the yarns to produce non-uniform random twist in the yarns.

Abstract: A light source includes a source of high frequency power, an electrodelsss lamp and a termination fixture for coupling power to the lamp, the fixture having an outer conductor and an inner conductor whose length is adjustable for providing dynamic matching of the impedance of the lamp during the operating mode to the output impedance of the source. Variations in the power level from the source, such as is desirable in providing lamp brightness selectivity, vary both the real and imaginary components of the lamp impedance, and corresponding adjustments of the inner conductor length provide a tuning technique to compensate for these variations.

Abstract: In a light source in which an electrodeless lamp is excited by high frequency power coupled into a lamp termination fixture having an inner and outer conductor, the fixture has a dynamically variable characteristic impedance, thereby facilitating dynamic matching of the variable impedance of the lamp during the lamp-off to lamp-run conditions to the output impedance of the power source. This feature is possible because the position of the inner conductor with respect to the outer conductor is controlled as a function of the heat generated within the fixture. The inner conductor is made of a conductive material having a non-uniform coefficient of thermal expansion. When the fixture is cold, the lamp and the lamp-coupled end of the inner conductor are substantially concentric to the outer conductor thereby creating a high characteristic impedance to match the high starting lamp impedance to the source impedance.